2019-09-26 09:12:36

by Anup Patel

Subject: Re: [PATCH 1/2] platform: goldfish: Allow goldfish virtual platform drivers for RISCV

On Wed, Sep 25, 2019 at 10:13 AM Greg Kroah-Hartman
<[email protected]> wrote:
>
> On Wed, Sep 25, 2019 at 04:30:03AM +0000, Anup Patel wrote:
> > We will be using some of the Goldfish virtual platform devices (such
> > as RTC) on QEMU RISC-V virt machine so this patch enables goldfish
> > kconfig option for RISC-V architecture.
> >
> > Signed-off-by: Anup Patel <[email protected]>
> > ---
> > drivers/platform/goldfish/Kconfig | 2 +-
> > 1 file changed, 1 insertion(+), 1 deletion(-)
> >
> > diff --git a/drivers/platform/goldfish/Kconfig b/drivers/platform/goldfish/Kconfig
> > index 77b35df3a801..0ba825030ffe 100644
> > --- a/drivers/platform/goldfish/Kconfig
> > +++ b/drivers/platform/goldfish/Kconfig
> > @@ -1,7 +1,7 @@
> > # SPDX-License-Identifier: GPL-2.0-only
> > menuconfig GOLDFISH
> > bool "Platform support for Goldfish virtual devices"
> > - depends on X86_32 || X86_64 || ARM || ARM64 || MIPS
> > + depends on X86_32 || X86_64 || ARM || ARM64 || MIPS || RISCV
>
> Why does this depend on any of these? Can't we just have:

May be Goldfish drivers were compile tested/tried on these architectures only.

>
> > depends on HAS_IOMEM
>
> And that's it?

I think it should be just "depends on HAS_IOMEM && HAS_DMA" just like
VirtIO MMIO. Agree ??

Regards,
Anup

2019-09-26 09:12:58

by Greg Kroah-Hartman

Subject: Re: [PATCH 1/2] platform: goldfish: Allow goldfish virtual platform drivers for RISCV

On Wed, Sep 25, 2019 at 10:30:00AM +0530, Anup Patel wrote:
> On Wed, Sep 25, 2019 at 10:13 AM Greg Kroah-Hartman
> <[email protected]> wrote:
> >
> > On Wed, Sep 25, 2019 at 04:30:03AM +0000, Anup Patel wrote:
> > > We will be using some of the Goldfish virtual platform devices (such
> > > as RTC) on QEMU RISC-V virt machine so this patch enables goldfish
> > > kconfig option for RISC-V architecture.
> > >
> > > Signed-off-by: Anup Patel <[email protected]>
> > > ---
> > > drivers/platform/goldfish/Kconfig | 2 +-
> > > 1 file changed, 1 insertion(+), 1 deletion(-)
> > >
> > > diff --git a/drivers/platform/goldfish/Kconfig b/drivers/platform/goldfish/Kconfig
> > > index 77b35df3a801..0ba825030ffe 100644
> > > --- a/drivers/platform/goldfish/Kconfig
> > > +++ b/drivers/platform/goldfish/Kconfig
> > > @@ -1,7 +1,7 @@
> > > # SPDX-License-Identifier: GPL-2.0-only
> > > menuconfig GOLDFISH
> > > bool "Platform support for Goldfish virtual devices"
> > > - depends on X86_32 || X86_64 || ARM || ARM64 || MIPS
> > > + depends on X86_32 || X86_64 || ARM || ARM64 || MIPS || RISCV
> >
> > Why does this depend on any of these? Can't we just have:
>
> May be Goldfish drivers were compile tested/tried on these architectures only.

True, but that does not mean a driver should only have a specific list
of arches. This should only be needed if you _know_ it doesn't work on
a specific arch, not the other way around.

> > > depends on HAS_IOMEM
> >
> > And that's it?
>
> I think it should be just "depends on HAS_IOMEM && HAS_DMA" just like
> VirtIO MMIO. Agree ??

No idea, but if that's what is needed for building, then sure :)

thanks,

greg k-h

2019-09-26 09:13:03

by Anup Patel

Subject: Re: [PATCH 1/2] platform: goldfish: Allow goldfish virtual platform drivers for RISCV

On Wed, Sep 25, 2019 at 10:37 AM Greg Kroah-Hartman
<[email protected]> wrote:
>
> On Wed, Sep 25, 2019 at 10:30:00AM +0530, Anup Patel wrote:
> > On Wed, Sep 25, 2019 at 10:13 AM Greg Kroah-Hartman
> > <[email protected]> wrote:
> > >
> > > On Wed, Sep 25, 2019 at 04:30:03AM +0000, Anup Patel wrote:
> > > > We will be using some of the Goldfish virtual platform devices (such
> > > > as RTC) on QEMU RISC-V virt machine so this patch enables goldfish
> > > > kconfig option for RISC-V architecture.
> > > >
> > > > Signed-off-by: Anup Patel <[email protected]>
> > > > ---
> > > > drivers/platform/goldfish/Kconfig | 2 +-
> > > > 1 file changed, 1 insertion(+), 1 deletion(-)
> > > >
> > > > diff --git a/drivers/platform/goldfish/Kconfig b/drivers/platform/goldfish/Kconfig
> > > > index 77b35df3a801..0ba825030ffe 100644
> > > > --- a/drivers/platform/goldfish/Kconfig
> > > > +++ b/drivers/platform/goldfish/Kconfig
> > > > @@ -1,7 +1,7 @@
> > > > # SPDX-License-Identifier: GPL-2.0-only
> > > > menuconfig GOLDFISH
> > > > bool "Platform support for Goldfish virtual devices"
> > > > - depends on X86_32 || X86_64 || ARM || ARM64 || MIPS
> > > > + depends on X86_32 || X86_64 || ARM || ARM64 || MIPS || RISCV
> > >
> > > Why does this depend on any of these? Can't we just have:
> >
> > May be Goldfish drivers were compile tested/tried on these architectures only.
>
> True, but that does not mean a driver should only have a specific list
> of arches. This should only be needed if you _know_ it doesn't work on
> a specific arch, not the other way around.

No problem, I will drop depends on various architectures line

>
> > > > depends on HAS_IOMEM
> > >
> > > And that's it?
> >
> > I think it should be just "depends on HAS_IOMEM && HAS_DMA" just like
> > VirtIO MMIO. Agree ??
>
> No idea, but if that's what is needed for building, then sure :)

The Goldfish framebuffer can do DMA access so I add dependency on
HAS_DMA.

Refer, https://android.googlesource.com/platform/external/qemu/+/master/docs/GOLDFISH-VIRTUAL-HARDWARE.TXT

I will send v2 as-per above.

Regards,
Anup