On 08/11/19 21:01, Andrea Arcangeli wrote: > On Fri, Nov 08, 2019 at 08:51:04PM +0100, Paolo Bonzini wrote: >> I suppose we could use code patching mechanism to avoid the retpolines. >> Andrea, what do you think about that? That would have the advantage >> that we won't have to remove kvm_x86_ops. :) > > page 17 covers pvops: > > https://people.redhat.com/~aarcange/slides/2019-KVM-monolithic.pdf You can patch call instructions directly using text_poke when kvm_intel.ko or kvm_amd.ko, I'm not sure why that would be worse for TLB or RAM usage. The hard part is recording the location of the call sites using some pushsection/popsection magic. Paolo
