KVM: arm64: Mark set_sysreg_masks() as inline to avoid build failure

When compiling with CONFIG_CC_OPTIMIZE_FOR_SIZE=y, set_sysreg_masks()
fails to compile thanks to:

	BUILD_BUG_ON(!__builtin_constant_p(sr));

as the compiler doesn't identify sr as a constant, despite all the
callers passing constants.

Fix the issue by always inlining this function, which allows GCC to
do the right thing.

Reported-by: kernel test robot <lkp@intel.com>
Closes: https://lore.kernel.org/oe-kbuild-all/202411201857.ZNudtGJl-lkp@intel.com/
Fixes: a0162020095e2 ("KVM: arm64: Extend masking facility to arbitrary registers")
Signed-off-by: Marc Zyngier <maz@kernel.org>
Reviewed-by: Joey Gouly <joey.gouly@arm.com>
Link: https://lore.kernel.org/r/20241120111516.304250-1-maz@kernel.org
Signed-off-by: Oliver Upton <oliver.upton@linux.dev>

diff --git a/arch/arm64/kvm/nested.c b/arch/arm64/kvm/nested.c
index aeaa6017ffd89f3a53938bb71b061e48d749d613..9b36218b48def495708f49b59e5f9211bd86cc0f 100644 (file)
--- a/arch/arm64/kvm/nested.c
+++ b/arch/arm64/kvm/nested.c
@@ -951,7 +951,7 @@ u64 kvm_vcpu_apply_reg_masks(const struct kvm_vcpu *vcpu,
        return v;
 }
 
-static void set_sysreg_masks(struct kvm *kvm, int sr, u64 res0, u64 res1)
+static __always_inline void set_sysreg_masks(struct kvm *kvm, int sr, u64 res0, u64 res1)
 {
        int i = sr - __SANITISED_REG_START__;