On Tue, Jun 30, 2026, Xiaoyao Li wrote:

On 6/19/2026 8:31 AM, Ackerley Tng via B4 Relay wrote:

quoted

  arch/x86/include/asm/kvm_host.h | 4 +++-
  include/linux/kvm_host.h        | 2 +-
  2 files changed, 4 insertions(+), 2 deletions(-)

diff --git a/arch/x86/include/asm/kvm_host.h b/arch/x86/include/asm/kvm_host.h
index 8e8eb8a5e8a6b..1bde67cf6eb0e 100644
--- a/arch/x86/include/asm/kvm_host.h
+++ b/arch/x86/include/asm/kvm_host.h

@@ -2394,7 +2394,9 @@ void kvm_configure_mmu(bool enable_tdp, int tdp_forced_root_level,
  		       int tdp_max_root_level, int tdp_huge_page_level);
-#ifdef CONFIG_KVM_VM_MEMORY_ATTRIBUTES
+#if defined(CONFIG_KVM_SW_PROTECTED_VM) ||	\
+	defined(CONFIG_KVM_INTEL_TDX) ||	\
+	defined(CONFIG_KVM_AMD_SEV)

Maybe we can just remove the #ifdef and make it always avaiable?

No, because common KVM keys off the macro to determine whether or not PRIVATE is
a supported attribute:

  #ifdef kvm_arch_has_private_mem
  static u64 kvm_supports_private_mem(struct kvm *kvm)
  {
	return !kvm || kvm_arch_has_private_mem(kvm);
  }
  #else
  #define kvm_supports_private_mem(kvm) false
  #endif

And also whether or not to provide the in-place conversion param (without PRIVATE,
conversions aren't supported in general):

  #ifdef kvm_arch_has_private_mem
  bool __ro_after_init gmem_in_place_conversion = !IS_ENABLED(CONFIG_KVM_VM_MEMORY_ATTRIBUTES);
  module_param(gmem_in_place_conversion, bool, 0444);
  EXPORT_SYMBOL_FOR_KVM_INTERNAL(gmem_in_place_conversion);
  #endif

I agree the #ifdeffery is ugly, but kvm_supports_private_mem() in particular
needs to evaluate to false if PRIVATE memory isn't supported.