| Message ID | 1621910615-29985-1-git-send-email-zhangshaokun@hisilicon.com (mailing list archive) |
|---|---|
| State | New, archived |
| Headers | show |
| Series | KVM: x86/mmu: Remove the repeated declaration | expand |
On Tue, May 25, 2021, Shaokun Zhang wrote: > Function 'is_nx_huge_page_enabled' is declared twice, remove the > repeated declaration. > > Cc: Ben Gardon <bgardon@google.com> > Cc: Paolo Bonzini <pbonzini@redhat.com> > Signed-off-by: Shaokun Zhang <zhangshaokun@hisilicon.com> > --- > arch/x86/kvm/mmu/mmu_internal.h | 2 -- > 1 file changed, 2 deletions(-) > > diff --git a/arch/x86/kvm/mmu/mmu_internal.h b/arch/x86/kvm/mmu/mmu_internal.h > index d64ccb417c60..54c6e6193ff2 100644 > --- a/arch/x86/kvm/mmu/mmu_internal.h > +++ b/arch/x86/kvm/mmu/mmu_internal.h > @@ -158,8 +158,6 @@ int kvm_mmu_hugepage_adjust(struct kvm_vcpu *vcpu, gfn_t gfn, > void disallowed_hugepage_adjust(u64 spte, gfn_t gfn, int cur_level, > kvm_pfn_t *pfnp, int *goal_levelp); > > -bool is_nx_huge_page_enabled(void); Rather than simply delete the extra declaration, what about converting this to static inline and opportunistically deleting the duplicate? The implementation is a single operation and this is MMU-internal, i.e. there's no need to export and limited exposure of nx_huge_pages to other code. > - > void *mmu_memory_cache_alloc(struct kvm_mmu_memory_cache *mc); > > void account_huge_nx_page(struct kvm *kvm, struct kvm_mmu_page *sp); > -- > 2.7.4 >
Hi Sean, On 2021/5/26 0:09, Sean Christopherson wrote: > On Tue, May 25, 2021, Shaokun Zhang wrote: >> Function 'is_nx_huge_page_enabled' is declared twice, remove the >> repeated declaration. >> >> Cc: Ben Gardon <bgardon@google.com> >> Cc: Paolo Bonzini <pbonzini@redhat.com> >> Signed-off-by: Shaokun Zhang <zhangshaokun@hisilicon.com> >> --- >> arch/x86/kvm/mmu/mmu_internal.h | 2 -- >> 1 file changed, 2 deletions(-) >> >> diff --git a/arch/x86/kvm/mmu/mmu_internal.h b/arch/x86/kvm/mmu/mmu_internal.h >> index d64ccb417c60..54c6e6193ff2 100644 >> --- a/arch/x86/kvm/mmu/mmu_internal.h >> +++ b/arch/x86/kvm/mmu/mmu_internal.h >> @@ -158,8 +158,6 @@ int kvm_mmu_hugepage_adjust(struct kvm_vcpu *vcpu, gfn_t gfn, >> void disallowed_hugepage_adjust(u64 spte, gfn_t gfn, int cur_level, >> kvm_pfn_t *pfnp, int *goal_levelp); >> >> -bool is_nx_huge_page_enabled(void); > > Rather than simply delete the extra declaration, what about converting this to > static inline and opportunistically deleting the duplicate? The implementation It seems that you want to make it static inline in mmu_internal.h, right? > is a single operation and this is MMU-internal, i.e. there's no need to export > and limited exposure of nx_huge_pages to other code. If is_nx_huge_page_enabled is inline in mmu_internal.h, nx_huge_pages will be external in mmu_internal.h and exposed to other code. Do I miss something? Thanks, Shaokun > >> - >> void *mmu_memory_cache_alloc(struct kvm_mmu_memory_cache *mc); >> >> void account_huge_nx_page(struct kvm *kvm, struct kvm_mmu_page *sp); >> -- >> 2.7.4 >> > . >
On Wed, May 26, 2021, Shaokun Zhang wrote: > Hi Sean, > > On 2021/5/26 0:09, Sean Christopherson wrote: > > Rather than simply delete the extra declaration, what about converting this to > > static inline and opportunistically deleting the duplicate? The implementation > > It seems that you want to make it static inline in mmu_internal.h, right? Yep. > > is a single operation and this is MMU-internal, i.e. there's no need to export > > and limited exposure of nx_huge_pages to other code. > > If is_nx_huge_page_enabled is inline in mmu_internal.h, nx_huge_pages will be > external in mmu_internal.h and exposed to other code. Do I miss something? Yes, it will be theoretically exposed to other code, but as the name suggests, mmu_internal.h should not be included by anything outside of mmu/, and of course we can reject patches that try to access nx_guest_pages directly.
diff --git a/arch/x86/kvm/mmu/mmu_internal.h b/arch/x86/kvm/mmu/mmu_internal.h index d64ccb417c60..54c6e6193ff2 100644 --- a/arch/x86/kvm/mmu/mmu_internal.h +++ b/arch/x86/kvm/mmu/mmu_internal.h @@ -158,8 +158,6 @@ int kvm_mmu_hugepage_adjust(struct kvm_vcpu *vcpu, gfn_t gfn, void disallowed_hugepage_adjust(u64 spte, gfn_t gfn, int cur_level, kvm_pfn_t *pfnp, int *goal_levelp); -bool is_nx_huge_page_enabled(void); - void *mmu_memory_cache_alloc(struct kvm_mmu_memory_cache *mc); void account_huge_nx_page(struct kvm *kvm, struct kvm_mmu_page *sp);
Function 'is_nx_huge_page_enabled' is declared twice, remove the repeated declaration. Cc: Ben Gardon <bgardon@google.com> Cc: Paolo Bonzini <pbonzini@redhat.com> Signed-off-by: Shaokun Zhang <zhangshaokun@hisilicon.com> --- arch/x86/kvm/mmu/mmu_internal.h | 2 -- 1 file changed, 2 deletions(-)