Re: [PATCH v16 4/4] m68k: introduce a virtual m68k machine

[Laurent Vivier]

Le 31/05/2022 à 03:14, Randy Dunlap a écrit :
> Hi--

Hi,

> On 4/6/22 13:15, Laurent Vivier wrote:
> > diff --git a/arch/m68k/Kconfig.machine b/arch/m68k/Kconfig.machine
> > index eeab4f3e6c19..188a8f8a0104 100644
> > --- a/arch/m68k/Kconfig.machine
> > +++ b/arch/m68k/Kconfig.machine
> > @@ -149,6 +149,23 @@ config SUN3
> >   
> >   	  If you don't want to compile a kernel exclusively for a Sun 3, say N.
> >   
> > +config VIRT
> > +	bool "Virtual M68k Machine support"
> > +	depends on MMU
> > +	select GENERIC_CLOCKEVENTS
> > +	select GOLDFISH
> > +	select GOLDFISH_TIMER
> > +	select GOLDFISH_TTY
> > +	select M68040
> > +	select MMU_MOTOROLA if MMU
> > +	select RTC_CLASS
> > +	select RTC_DRV_GOLDFISH
> > +	select TTY
> > +	select VIRTIO_MMIO
>
> This select VIRTIO_MMIO can cause a kconfig warning:
>
> WARNING: unmet direct dependencies detected for VIRTIO_MMIO
>    Depends on [n]: VIRTIO_MENU [=n] && HAS_IOMEM [=y] && HAS_DMA [=y]

In my repo, VIRTIO_MMIO has no dependency on VIRTIO_MENU:

./drivers/virtio/Kconfig
config VIRTIO_MMIO
        tristate "Platform bus driver for memory mapped virtio devices"
        depends on HAS_IOMEM && HAS_DMA

>    Selected by [y]:
>    - VIRT [=y] && M68KCLASSIC [=y] && MMU [=y]

How do you generate this warning?

The first version of my patch added the VIRTIO_MENU but Geert said it's not necessary because 
VIRTIO_MENU defaults to y.

https://lore.kernel.org/lkml/CAMuHMdUFh2W-bY5Ez1aOTZQjq0=THvmOf22JdxWoNNtFLskSzw@xxxxxxxxxxxxxx/

> > +	help
> > +	  This options enable a pure virtual machine based on m68k,
> > +	  VIRTIO MMIO devices and GOLDFISH interfaces (TTY, RTC, PIC)
>
> The sentence above needs an ending period ('.').

I agree.

Thanks,
Laurent