powerpc/kvm: Handle transparent hugepage in KVM

We can find pte that are splitting while walking page tables. Return
None pte in that case.

Signed-off-by: Aneesh Kumar K.V <aneesh.kumar@linux.vnet.ibm.com>
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>

3 files changed, 44 insertions, 34 deletions

diff --git a/arch/powerpc/include/asm/kvm_book3s_64.h b/arch/powerpc/include/asm/kvm_book3s_64.h
index 9c1ff330c805..a1ecb14e4442 100644
--- a/arch/powerpc/include/asm/kvm_book3s_64.h
+++ b/arch/powerpc/include/asm/kvm_book3s_64.h
@@ -159,36 +159,46 @@ static inline int hpte_cache_flags_ok(unsigned long ptel, unsigned long io_type)
 }
 
 /*
- * Lock and read a linux PTE.  If it's present and writable, atomically
- * set dirty and referenced bits and return the PTE, otherwise return 0.
+ * If it's present and writable, atomically set dirty and referenced bits and
+ * return the PTE, otherwise return 0. If we find a transparent hugepage
+ * and if it is marked splitting we return 0;
  */
-static inline pte_t kvmppc_read_update_linux_pte(pte_t *p, int writing)
+static inline pte_t kvmppc_read_update_linux_pte(pte_t *ptep, int writing,
+						 unsigned int hugepage)
 {
-	pte_t pte, tmp;
-
-	/* wait until _PAGE_BUSY is clear then set it atomically */
-	__asm__ __volatile__ (
-		"1:	ldarx	%0,0,%3\n"
-		"	andi.	%1,%0,%4\n"
-		"	bne-	1b\n"
-		"	ori	%1,%0,%4\n"
-		"	stdcx.	%1,0,%3\n"
-		"	bne-	1b"
-		: "=&r" (pte), "=&r" (tmp), "=m" (*p)
-		: "r" (p), "i" (_PAGE_BUSY)
-		: "cc");
-
-	if (pte_present(pte)) {
-		pte = pte_mkyoung(pte);
-		if (writing && pte_write(pte))
-			pte = pte_mkdirty(pte);
-	}
+	pte_t old_pte, new_pte = __pte(0);
+
+	while (1) {
+		old_pte = pte_val(*ptep);
+		/*
+		 * wait until _PAGE_BUSY is clear then set it atomically
+		 */
+		if (unlikely(old_pte & _PAGE_BUSY)) {
+			cpu_relax();
+			continue;
+		}
+#ifdef CONFIG_TRANSPARENT_HUGEPAGE
+		/* If hugepage and is trans splitting return None */
+		if (unlikely(hugepage &&
+			     pmd_trans_splitting(pte_pmd(old_pte))))
+			return __pte(0);
+#endif
+		/* If pte is not present return None */
+		if (unlikely(!(old_pte & _PAGE_PRESENT)))
+			return __pte(0);
 
-	*p = pte;	/* clears _PAGE_BUSY */
+		new_pte = pte_mkyoung(old_pte);
+		if (writing && pte_write(old_pte))
+			new_pte = pte_mkdirty(new_pte);
 
-	return pte;
+		if (old_pte == __cmpxchg_u64((unsigned long *)ptep, old_pte,
+					     new_pte))
+			break;
+	}
+	return new_pte;
 }
 
+
 /* Return HPTE cache control bits corresponding to Linux pte bits */
 static inline unsigned long hpte_cache_bits(unsigned long pte_val)
 {
diff --git a/arch/powerpc/kvm/book3s_64_mmu_hv.c b/arch/powerpc/kvm/book3s_64_mmu_hv.c
index 5880dfb31074..710d31317d81 100644
--- a/arch/powerpc/kvm/book3s_64_mmu_hv.c
+++ b/arch/powerpc/kvm/book3s_64_mmu_hv.c
@@ -675,6 +675,7 @@ int kvmppc_book3s_hv_page_fault(struct kvm_run *run, struct kvm_vcpu *vcpu,
 		}
 		/* if the guest wants write access, see if that is OK */
 		if (!writing && hpte_is_writable(r)) {
+			unsigned int hugepage_shift;
 			pte_t *ptep, pte;
 
 			/*
@@ -683,9 +684,10 @@ int kvmppc_book3s_hv_page_fault(struct kvm_run *run, struct kvm_vcpu *vcpu,
 			 */
 			rcu_read_lock_sched();
 			ptep = find_linux_pte_or_hugepte(current->mm->pgd,
-							 hva, NULL);
-			if (ptep && pte_present(*ptep)) {
-				pte = kvmppc_read_update_linux_pte(ptep, 1);
+							 hva, &hugepage_shift);
+			if (ptep) {
+				pte = kvmppc_read_update_linux_pte(ptep, 1,
+							   hugepage_shift);
 				if (pte_write(pte))
 					write_ok = 1;
 			}
diff --git a/arch/powerpc/kvm/book3s_hv_rm_mmu.c b/arch/powerpc/kvm/book3s_hv_rm_mmu.c
index dcf892d25a56..fc25689a9f35 100644
--- a/arch/powerpc/kvm/book3s_hv_rm_mmu.c
+++ b/arch/powerpc/kvm/book3s_hv_rm_mmu.c
@@ -139,20 +139,18 @@ static pte_t lookup_linux_pte(pgd_t *pgdir, unsigned long hva,
 {
 	pte_t *ptep;
 	unsigned long ps = *pte_sizep;
-	unsigned int shift;
+	unsigned int hugepage_shift;
 
-	ptep = find_linux_pte_or_hugepte(pgdir, hva, &shift);
+	ptep = find_linux_pte_or_hugepte(pgdir, hva, &hugepage_shift);
 	if (!ptep)
 		return __pte(0);
-	if (shift)
-		*pte_sizep = 1ul << shift;
+	if (hugepage_shift)
+		*pte_sizep = 1ul << hugepage_shift;
 	else
 		*pte_sizep = PAGE_SIZE;
 	if (ps > *pte_sizep)
 		return __pte(0);
-	if (!pte_present(*ptep))
-		return __pte(0);
-	return kvmppc_read_update_linux_pte(ptep, writing);
+	return kvmppc_read_update_linux_pte(ptep, writing, hugepage_shift);
 }
 
 static inline void unlock_hpte(unsigned long *hpte, unsigned long hpte_v)