On 5/22/2026 7:16 AM, Lisa Wang wrote: > Force GUEST_MEMFD for the primary memory region of TDX VMs. > > TDX must use guest_memfd for private pages as there is no alternative > mechanism supported by the TDX architecture.
This sounds a bit confusing to me. It's the kernel/KVM only support guest_memfd for private pages. The restriction is not from the "TDX architecture". Also, the short log is also a bit confusing. Why describe the "first memory region" here? > > Signed-off-by: Lisa Wang <[email protected]> > --- > tools/testing/selftests/kvm/lib/kvm_util.c | 2 +- > 1 file changed, 1 insertion(+), 1 deletion(-) > > diff --git a/tools/testing/selftests/kvm/lib/kvm_util.c > b/tools/testing/selftests/kvm/lib/kvm_util.c > index d1befa3f4b30..9a29540fff40 100644 > --- a/tools/testing/selftests/kvm/lib/kvm_util.c > +++ b/tools/testing/selftests/kvm/lib/kvm_util.c > @@ -472,7 +472,7 @@ void kvm_set_files_rlimit(u32 nr_vcpus) > static bool is_guest_memfd_required(struct vm_shape shape) > { > #ifdef __x86_64__ > - return shape.type == KVM_X86_SNP_VM; > + return (shape.type == KVM_X86_SNP_VM || shape.type == KVM_X86_TDX_VM); > #else > return false; > #endif >
