| Message ID | 20240423191728.52535-1-pbonzini@redhat.com (mailing list archive) |
|---|---|
| State | New |
| Headers | show |
| Series | Kconfig: kvm: allow building without any board | expand |
On 23/4/24 21:17, Paolo Bonzini wrote: > KVM code might have to call functions on the PCIDevice that is > passed to kvm_arch_fixup_msi_route(). This fails in the case > where --without-default-devices is used and no board is > configured. While this is not really a useful configuration, > and therefore setting up stubs for CONFIG_PCI is overkill, > failing the build is impolite. Just include the PCI > subsystem if kvm_arch_fixup_msi_route() requires it, as > is the case for ARM and x86. > > Reported-by: Philippe Mathieu-Daudé <philmd@linaro.org> > Tested-by: Fabiano Rosas <farosas@suse.de> > Signed-off-by: Paolo Bonzini <pbonzini@redhat.com> > --- > target/arm/Kconfig | 2 ++ > target/i386/Kconfig | 2 ++ > 2 files changed, 4 insertions(+) > > diff --git a/target/arm/Kconfig b/target/arm/Kconfig > index bf57d739cd1..5847c5a74a7 100644 > --- a/target/arm/Kconfig > +++ b/target/arm/Kconfig > @@ -9,3 +9,5 @@ config ARM > config AARCH64 > bool > select ARM > + # kvm_arch_fixup_msi_route() needs to access PCIDevice > + select PCI if KVM Kind of funny :) Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org>
diff --git a/target/arm/Kconfig b/target/arm/Kconfig index bf57d739cd1..5847c5a74a7 100644 --- a/target/arm/Kconfig +++ b/target/arm/Kconfig @@ -9,3 +9,5 @@ config ARM config AARCH64 bool select ARM + # kvm_arch_fixup_msi_route() needs to access PCIDevice + select PCI if KVM diff --git a/target/i386/Kconfig b/target/i386/Kconfig index ce6968906ee..46898946394 100644 --- a/target/i386/Kconfig +++ b/target/i386/Kconfig @@ -1,5 +1,7 @@ config I386 bool + # kvm_arch_fixup_msi_route() needs to access PCIDevice + select PCI if KVM config X86_64 bool