On 9/4/19 1:32 AM, Andrew Jones wrote: > Should we already introduce a function that will collect all > finalizers together now, rather than sprinkling around > arm_cpu_sve_finalize() calls? Something like > > void arm_cpu_finalize_features(ARMCPU *cpu, Error **errp) > { > arm_cpu_sve_finalize(cpu, errp); > } > > Of course we can introduce it when/if we add other finalizers > later, but I guess the vfp-neon finalizer should be coming > soon anyway.
That sounds reasonable. r~