| Message ID | 20241022085359.186470-1-pbonzini@redhat.com (mailing list archive) |
|---|---|
| State | New |
| Headers | show |
| Series | arm: Kconfig: disable stellaris if Rust is enabled | expand |
On Tue, 22 Oct 2024 at 09:54, Paolo Bonzini <pbonzini@redhat.com> wrote: > > The stellaris board requires the pl011-luminary variant of pl011, > which is not supported by the Rust implementation. > > There are at least three possibilities: 1) implement the subclass > (a bit harder in Rust since the language does not have subclasses) > 2) change the ID to a property 3) split pl011-luminary to a separate > Kconfig symbol and leave the subclass as C code. > > Just for the sake of starting the discussion, this RFC patch uses > the big axe and disables stellaris. Reasonable enough as a starter. For the longer term, we should figure out how we want to handle this kind of "we have a common base QOM class that implements almost all of the behaviour and we subclass it to implement different minor variants" in Rust, because we use it a fair amount in the existing C code. -- PMM
Hello Paolo, On Tue, 22 Oct 2024 at 11:54, Paolo Bonzini <pbonzini@redhat.com> wrote: > > The stellaris board requires the pl011-luminary variant of pl011, > which is not supported by the Rust implementation. > > There are at least three possibilities: 1) implement the subclass > (a bit harder in Rust since the language does not have subclasses) > 2) change the ID to a property 3) split pl011-luminary to a separate > Kconfig symbol and leave the subclass as C code. > > Just for the sake of starting the discussion, this RFC patch uses > the big axe and disables stellaris. > > Signed-off-by: Paolo Bonzini <pbonzini@redhat.com> > --- > hw/arm/Kconfig | 4 ++-- > 1 file changed, 2 insertions(+), 2 deletions(-) > > diff --git a/hw/arm/Kconfig b/hw/arm/Kconfig > index e7fd9338d11..d5ade150d23 100644 > --- a/hw/arm/Kconfig > +++ b/hw/arm/Kconfig > @@ -219,12 +219,12 @@ config STELLARIS > bool > default y > depends on TCG && ARM > + depends on !HAVE_RUST # Rust does not implement pl011-luminary > imply I2C_DEVICES > select ARM_V7M > select CMSDK_APB_WATCHDOG > select I2C > - select PL011 if !HAVE_RUST # UART > - select X_PL011_RUST if HAVE_RUST # UART > + select PL011 # UART > select PL022 # SPI > select PL061 # GPIO > select SSD0303 # OLED display > -- > 2.46.2 > Reviewed-by: Manos Pitsidianakis <manos.pitsidianakis@linaro.org> By the way I already have a patch that adds a -luminary version, planning on including it with the rest of my procedural macro patches I will send.
On Wed, Oct 23, 2024 at 12:28 PM Manos Pitsidianakis <manos.pitsidianakis@linaro.org> wrote: > By the way I already have a patch that adds a -luminary version, > planning on including it with the rest of my procedural macro patches > I will send. I'll hold on this patch until I get yours then. It is not needed yet for CI, since we don't run tests with Rust enabled. Paolo
diff --git a/hw/arm/Kconfig b/hw/arm/Kconfig index e7fd9338d11..d5ade150d23 100644 --- a/hw/arm/Kconfig +++ b/hw/arm/Kconfig @@ -219,12 +219,12 @@ config STELLARIS bool default y depends on TCG && ARM + depends on !HAVE_RUST # Rust does not implement pl011-luminary imply I2C_DEVICES select ARM_V7M select CMSDK_APB_WATCHDOG select I2C - select PL011 if !HAVE_RUST # UART - select X_PL011_RUST if HAVE_RUST # UART + select PL011 # UART select PL022 # SPI select PL061 # GPIO select SSD0303 # OLED display
The stellaris board requires the pl011-luminary variant of pl011, which is not supported by the Rust implementation. There are at least three possibilities: 1) implement the subclass (a bit harder in Rust since the language does not have subclasses) 2) change the ID to a property 3) split pl011-luminary to a separate Kconfig symbol and leave the subclass as C code. Just for the sake of starting the discussion, this RFC patch uses the big axe and disables stellaris. Signed-off-by: Paolo Bonzini <pbonzini@redhat.com> --- hw/arm/Kconfig | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-)