> > Further, considering that this is currently the only case that needs to > > to set the VM level's capability in the CPU context, there is no need to > > introduce a new kvm interface (in your previous patch), which can instead > > be set in kvm_cpu_realizefn(), like: > > Now your case is not the only user of kvm_arch_pre_create_vcpu(), and TDX also needs this [*]. So, this is the support for bringing back your previous solution (preferably with comments, as I suggested earlier, explaining why it's necessary to handle VM-level cap in the CPU context). :-) [*]: https://lore.kernel.org/qemu-devel/20241105062408.3533704-8-xiaoyao.li@xxxxxxxxx/
