One quick comment while it's on my mind, I'll give this a proper gander tomorrow. On Tue, Feb 02, 2021, Michael Roth wrote: > diff --git a/arch/x86/kvm/svm/svm_ops.h b/arch/x86/kvm/svm/svm_ops.h > index 0c8377aee52c..c2a05f56c8e4 100644 > --- a/arch/x86/kvm/svm/svm_ops.h > +++ b/arch/x86/kvm/svm/svm_ops.h > @@ -56,4 +56,9 @@ static inline void vmsave(hpa_t pa) > svm_asm1(vmsave, "a" (pa), "memory"); > } > > +static inline void vmload(hpa_t pa) This needs to be 'unsigned long', using 'hpa_t' in vmsave() is wrong as the instructions consume rAX based on effective address. I wrote the function comment for the vmsave() fix so that it applies to both VMSAVE and VMLOAD, so this can be a simple fixup on application (assuming v5 isn't needed for other reasons). https://lkml.kernel.org/r/20210202223416.2702336-1-seanjc@xxxxxxxxxx > +{ > + svm_asm1(vmload, "a" (pa), "memory"); > +} > + > #endif /* __KVM_X86_SVM_OPS_H */ > -- > 2.25.1 >
