On Tue, Feb 27, 2024 at 06:20:50PM -0500, Paolo Bonzini wrote:
> From: Isaku Yamahata <isaku.yamahata@intel.com>
>
> Refactor tdp_mmu_alloc_sp() and tdp_mmu_init_sp and eliminate
^
tdp_mmu_init_sp()
> tdp_mmu_init_child_sp(). Currently tdp_mmu_init_sp() (or
> tdp_mmu_init_child_sp()) sets kvm_mmu_page.role after tdp_mmu_alloc_sp()
> allocating struct kvm_mmu_page and its page table page. This patch makes
> tdp_mmu_alloc_sp() initialize kvm_mmu_page.role instead of
> tdp_mmu_init_sp().
>
> To handle private page tables, argument of is_private needs to be passed
> down. Given that already page level is passed down, it would be cumbersome
> to add one more parameter about sp. Instead replace the level argument with
> union kvm_mmu_page_role. Thus the number of argument won't be increased
This section is hard to understand. I'm lost at which functions are
mentioned here that took the level argument and should be replaced by
role.
> and more info about sp can be passed down.
My understanding of the change is:
Extra handling is need for Allocation of private page tables, so
earlier caculate the kvm_mmu_page_role for the sp and pass it to
tdp_mmu_alloc_sp(). Since the sp.role could be decided on sp
allocation, in turn remove the role argument for tdp_mmu_init_sp(), also
eliminate the helper tdp_mmu_init_child_sp().
>
> For private sp, secure page table will be also allocated in addition to
> struct kvm_mmu_page and page table (spt member). The allocation functions
> (tdp_mmu_alloc_sp() and __tdp_mmu_alloc_sp_for_split()) need to know if the
> allocation is for the conventional page table or private page table. Pass
> union kvm_mmu_role to those functions and initialize role member of struct
^
Should be kvm_mmu_page_role
Thanks,
Yilun
@@ -135,4 +135,16 @@ void tdp_iter_start(struct tdp_iter *iter, struct kvm_mmu_page *root,
void tdp_iter_next(struct tdp_iter *iter);
void tdp_iter_restart(struct tdp_iter *iter);
+static inline union kvm_mmu_page_role tdp_iter_child_role(struct tdp_iter *iter)
+{
+ union kvm_mmu_page_role child_role;
+ struct kvm_mmu_page *parent_sp;
+
+ parent_sp = sptep_to_sp(rcu_dereference(iter->sptep));
+
+ child_role = parent_sp->role;
+ child_role.level--;
+ return child_role;
+}
+
#endif /* __KVM_X86_MMU_TDP_ITER_H */
@@ -184,24 +184,30 @@ static struct kvm_mmu_page *tdp_mmu_next_root(struct kvm *kvm,
#define for_each_valid_tdp_mmu_root(_kvm, _root, _as_id) \
__for_each_tdp_mmu_root(_kvm, _root, _as_id, true)
-static struct kvm_mmu_page *tdp_mmu_alloc_sp(struct kvm_vcpu *vcpu)
+static struct kvm_mmu_page *tdp_mmu_alloc_sp(struct kvm_vcpu *vcpu,
+ union kvm_mmu_page_role role)
{
struct kvm_mmu_page *sp;
sp = kvm_mmu_memory_cache_alloc(&vcpu->arch.mmu_page_header_cache);
sp->spt = kvm_mmu_memory_cache_alloc(&vcpu->arch.mmu_shadow_page_cache);
+ sp->role = role;
return sp;
}
static void tdp_mmu_init_sp(struct kvm_mmu_page *sp, tdp_ptep_t sptep,
- gfn_t gfn, union kvm_mmu_page_role role)
+ gfn_t gfn)
{
INIT_LIST_HEAD(&sp->possible_nx_huge_page_link);
set_page_private(virt_to_page(sp->spt), (unsigned long)sp);
- sp->role = role;
+ /*
+ * role must be set before calling this function. At least role.level
+ * is not 0 (PG_LEVEL_NONE).
+ */
+ WARN_ON_ONCE(!sp->role.word);
sp->gfn = gfn;
sp->ptep = sptep;
sp->tdp_mmu_page = true;
@@ -209,20 +215,6 @@ static void tdp_mmu_init_sp(struct kvm_mmu_page *sp, tdp_ptep_t sptep,
trace_kvm_mmu_get_page(sp, true);
}
-static void tdp_mmu_init_child_sp(struct kvm_mmu_page *child_sp,
- struct tdp_iter *iter)
-{
- struct kvm_mmu_page *parent_sp;
- union kvm_mmu_page_role role;
-
- parent_sp = sptep_to_sp(rcu_dereference(iter->sptep));
-
- role = parent_sp->role;
- role.level--;
-
- tdp_mmu_init_sp(child_sp, iter->sptep, iter->gfn, role);
-}
-
int kvm_tdp_mmu_alloc_root(struct kvm_vcpu *vcpu)
{
struct kvm_mmu *mmu = vcpu->arch.mmu;
@@ -260,8 +252,8 @@ int kvm_tdp_mmu_alloc_root(struct kvm_vcpu *vcpu)
goto out_spin_unlock;
}
- root = tdp_mmu_alloc_sp(vcpu);
- tdp_mmu_init_sp(root, NULL, 0, role);
+ root = tdp_mmu_alloc_sp(vcpu, role);
+ tdp_mmu_init_sp(root, NULL, 0);
/*
* TDP MMU roots are kept until they are explicitly invalidated, either
@@ -1118,8 +1110,8 @@ int kvm_tdp_mmu_map(struct kvm_vcpu *vcpu, struct kvm_page_fault *fault)
* The SPTE is either non-present or points to a huge page that
* needs to be split.
*/
- sp = tdp_mmu_alloc_sp(vcpu);
- tdp_mmu_init_child_sp(sp, &iter);
+ sp = tdp_mmu_alloc_sp(vcpu, tdp_iter_child_role(&iter));
+ tdp_mmu_init_sp(sp, iter.sptep, iter.gfn);
sp->nx_huge_page_disallowed = fault->huge_page_disallowed;
@@ -1362,7 +1354,7 @@ bool kvm_tdp_mmu_wrprot_slot(struct kvm *kvm,
return spte_set;
}
-static struct kvm_mmu_page *__tdp_mmu_alloc_sp_for_split(gfp_t gfp)
+static struct kvm_mmu_page *__tdp_mmu_alloc_sp_for_split(gfp_t gfp, union kvm_mmu_page_role role)
{
struct kvm_mmu_page *sp;
@@ -1372,6 +1364,7 @@ static struct kvm_mmu_page *__tdp_mmu_alloc_sp_for_split(gfp_t gfp)
if (!sp)
return NULL;
+ sp->role = role;
sp->spt = (void *)__get_free_page(gfp);
if (!sp->spt) {
kmem_cache_free(mmu_page_header_cache, sp);
@@ -1385,6 +1378,7 @@ static struct kvm_mmu_page *tdp_mmu_alloc_sp_for_split(struct kvm *kvm,
struct tdp_iter *iter,
bool shared)
{
+ union kvm_mmu_page_role role = tdp_iter_child_role(iter);
struct kvm_mmu_page *sp;
kvm_lockdep_assert_mmu_lock_held(kvm, shared);
@@ -1398,7 +1392,7 @@ static struct kvm_mmu_page *tdp_mmu_alloc_sp_for_split(struct kvm *kvm,
* If this allocation fails we drop the lock and retry with reclaim
* allowed.
*/
- sp = __tdp_mmu_alloc_sp_for_split(GFP_NOWAIT | __GFP_ACCOUNT);
+ sp = __tdp_mmu_alloc_sp_for_split(GFP_NOWAIT | __GFP_ACCOUNT, role);
if (sp)
return sp;
@@ -1410,7 +1404,7 @@ static struct kvm_mmu_page *tdp_mmu_alloc_sp_for_split(struct kvm *kvm,
write_unlock(&kvm->mmu_lock);
iter->yielded = true;
- sp = __tdp_mmu_alloc_sp_for_split(GFP_KERNEL_ACCOUNT);
+ sp = __tdp_mmu_alloc_sp_for_split(GFP_KERNEL_ACCOUNT, role);
if (shared)
read_lock(&kvm->mmu_lock);
@@ -1505,7 +1499,7 @@ static int tdp_mmu_split_huge_pages_root(struct kvm *kvm,
continue;
}
- tdp_mmu_init_child_sp(sp, &iter);
+ tdp_mmu_init_sp(sp, iter.sptep, iter.gfn);
if (tdp_mmu_split_huge_page(kvm, &iter, sp, shared))
goto retry;