diff --git a/arch/arm/include/asm/spinlock.h b/arch/arm/include/asm/spinlock.h
index ac4bfae26702..0fa418463f49 100644
--- a/arch/arm/include/asm/spinlock.h
+++ b/arch/arm/include/asm/spinlock.h
@@ -120,12 +120,12 @@ static inline int arch_spin_value_unlocked(arch_spinlock_t lock)
 
 static inline int arch_spin_is_locked(arch_spinlock_t *lock)
 {
-	return !arch_spin_value_unlocked(ACCESS_ONCE(*lock));
+	return !arch_spin_value_unlocked(READ_ONCE(*lock));
 }
 
 static inline int arch_spin_is_contended(arch_spinlock_t *lock)
 {
-	struct __raw_tickets tickets = ACCESS_ONCE(lock->tickets);
+	struct __raw_tickets tickets = READ_ONCE(lock->tickets);
 	return (tickets.next - tickets.owner) > 1;
 }
 #define arch_spin_is_contended	arch_spin_is_contended
diff --git a/arch/arm64/include/asm/spinlock.h b/arch/arm64/include/asm/spinlock.h
index c45b7b1b7197..cee128732435 100644
--- a/arch/arm64/include/asm/spinlock.h
+++ b/arch/arm64/include/asm/spinlock.h
@@ -99,12 +99,12 @@ static inline int arch_spin_value_unlocked(arch_spinlock_t lock)
 
 static inline int arch_spin_is_locked(arch_spinlock_t *lock)
 {
-	return !arch_spin_value_unlocked(ACCESS_ONCE(*lock));
+	return !arch_spin_value_unlocked(READ_ONCE(*lock));
 }
 
 static inline int arch_spin_is_contended(arch_spinlock_t *lock)
 {
-	arch_spinlock_t lockval = ACCESS_ONCE(*lock);
+	arch_spinlock_t lockval = READ_ONCE(*lock);
 	return (lockval.next - lockval.owner) > 1;
 }
 #define arch_spin_is_contended	arch_spin_is_contended
diff --git a/arch/mips/mm/gup.c b/arch/mips/mm/gup.c
index 7cba480568c8..70795a67a276 100644
--- a/arch/mips/mm/gup.c
+++ b/arch/mips/mm/gup.c
@@ -30,7 +30,7 @@ retry:
 
 	return pte;
 #else
-	return ACCESS_ONCE(*ptep);
+	return READ_ONCE(*ptep);
 #endif
 }
 
diff --git a/arch/s390/kvm/gaccess.c b/arch/s390/kvm/gaccess.c
index 8b9ccf02a2c5..8a1be9017730 100644
--- a/arch/s390/kvm/gaccess.c
+++ b/arch/s390/kvm/gaccess.c
@@ -227,12 +227,10 @@ static void ipte_lock_simple(struct kvm_vcpu *vcpu)
 		goto out;
 	ic = &vcpu->kvm->arch.sca->ipte_control;
 	do {
-		old = *ic;
-		barrier();
+		old = READ_ONCE(*ic);
 		while (old.k) {
 			cond_resched();
-			old = *ic;
-			barrier();
+			old = READ_ONCE(*ic);
 		}
 		new = old;
 		new.k = 1;
@@ -251,8 +249,7 @@ static void ipte_unlock_simple(struct kvm_vcpu *vcpu)
 		goto out;
 	ic = &vcpu->kvm->arch.sca->ipte_control;
 	do {
-		old = *ic;
-		barrier();
+		old = READ_ONCE(*ic);
 		new = old;
 		new.k = 0;
 	} while (cmpxchg(&ic->val, old.val, new.val) != old.val);
@@ -267,12 +264,10 @@ static void ipte_lock_siif(struct kvm_vcpu *vcpu)
 
 	ic = &vcpu->kvm->arch.sca->ipte_control;
 	do {
-		old = *ic;
-		barrier();
+		old = READ_ONCE(*ic);
 		while (old.kg) {
 			cond_resched();
-			old = *ic;
-			barrier();
+			old = READ_ONCE(*ic);
 		}
 		new = old;
 		new.k = 1;
@@ -286,8 +281,7 @@ static void ipte_unlock_siif(struct kvm_vcpu *vcpu)
 
 	ic = &vcpu->kvm->arch.sca->ipte_control;
 	do {
-		old = *ic;
-		barrier();
+		old = READ_ONCE(*ic);
 		new = old;
 		new.kh--;
 		if (!new.kh)
diff --git a/arch/x86/include/asm/spinlock.h b/arch/x86/include/asm/spinlock.h
index a4efe477ceab..625660f8a2fc 100644
--- a/arch/x86/include/asm/spinlock.h
+++ b/arch/x86/include/asm/spinlock.h
@@ -92,7 +92,7 @@ static __always_inline void arch_spin_lock(arch_spinlock_t *lock)
 		unsigned count = SPIN_THRESHOLD;
 
 		do {
-			if (ACCESS_ONCE(lock->tickets.head) == inc.tail)
+			if (READ_ONCE(lock->tickets.head) == inc.tail)
 				goto out;
 			cpu_relax();
 		} while (--count);
@@ -105,7 +105,7 @@ static __always_inline int arch_spin_trylock(arch_spinlock_t *lock)
 {
 	arch_spinlock_t old, new;
 
-	old.tickets = ACCESS_ONCE(lock->tickets);
+	old.tickets = READ_ONCE(lock->tickets);
 	if (old.tickets.head != (old.tickets.tail & ~TICKET_SLOWPATH_FLAG))
 		return 0;
 
@@ -162,14 +162,14 @@ static __always_inline void arch_spin_unlock(arch_spinlock_t *lock)
 
 static inline int arch_spin_is_locked(arch_spinlock_t *lock)
 {
-	struct __raw_tickets tmp = ACCESS_ONCE(lock->tickets);
+	struct __raw_tickets tmp = READ_ONCE(lock->tickets);
 
 	return tmp.tail != tmp.head;
 }
 
 static inline int arch_spin_is_contended(arch_spinlock_t *lock)
 {
-	struct __raw_tickets tmp = ACCESS_ONCE(lock->tickets);
+	struct __raw_tickets tmp = READ_ONCE(lock->tickets);
 
 	return (__ticket_t)(tmp.tail - tmp.head) > TICKET_LOCK_INC;
 }
diff --git a/arch/x86/mm/gup.c b/arch/x86/mm/gup.c
index 207d9aef662d..d7547824e763 100644
--- a/arch/x86/mm/gup.c
+++ b/arch/x86/mm/gup.c
@@ -15,7 +15,7 @@
 static inline pte_t gup_get_pte(pte_t *ptep)
 {
 #ifndef CONFIG_X86_PAE
-	return ACCESS_ONCE(*ptep);
+	return READ_ONCE(*ptep);
 #else
 	/*
 	 * With get_user_pages_fast, we walk down the pagetables without taking
diff --git a/include/linux/compiler.h b/include/linux/compiler.h
index d5ad7b1118fc..a1c81f80978e 100644
--- a/include/linux/compiler.h
+++ b/include/linux/compiler.h
@@ -186,6 +186,80 @@ void ftrace_likely_update(struct ftrace_branch_data *f, int val, int expect);
 # define __UNIQUE_ID(prefix) __PASTE(__PASTE(__UNIQUE_ID_, prefix), __LINE__)
 #endif
 
+#include <uapi/linux/types.h>
+
+static __always_inline void data_access_exceeds_word_size(void)
+#ifdef __compiletime_warning
+__compiletime_warning("data access exceeds word size and won't be atomic")
+#endif
+;
+
+static __always_inline void data_access_exceeds_word_size(void)
+{
+}
+
+static __always_inline void __read_once_size(volatile void *p, void *res, int size)
+{
+	switch (size) {
+	case 1: *(__u8 *)res = *(volatile __u8 *)p; break;
+	case 2: *(__u16 *)res = *(volatile __u16 *)p; break;
+	case 4: *(__u32 *)res = *(volatile __u32 *)p; break;
+#ifdef CONFIG_64BIT
+	case 8: *(__u64 *)res = *(volatile __u64 *)p; break;
+#endif
+	default:
+		barrier();
+		__builtin_memcpy((void *)res, (const void *)p, size);
+		data_access_exceeds_word_size();
+		barrier();
+	}
+}
+
+static __always_inline void __assign_once_size(volatile void *p, void *res, int size)
+{
+	switch (size) {
+	case 1: *(volatile __u8 *)p = *(__u8 *)res; break;
+	case 2: *(volatile __u16 *)p = *(__u16 *)res; break;
+	case 4: *(volatile __u32 *)p = *(__u32 *)res; break;
+#ifdef CONFIG_64BIT
+	case 8: *(volatile __u64 *)p = *(__u64 *)res; break;
+#endif
+	default:
+		barrier();
+		__builtin_memcpy((void *)p, (const void *)res, size);
+		data_access_exceeds_word_size();
+		barrier();
+	}
+}
+
+/*
+ * Prevent the compiler from merging or refetching reads or writes. The
+ * compiler is also forbidden from reordering successive instances of
+ * READ_ONCE, ASSIGN_ONCE and ACCESS_ONCE (see below), but only when the
+ * compiler is aware of some particular ordering.  One way to make the
+ * compiler aware of ordering is to put the two invocations of READ_ONCE,
+ * ASSIGN_ONCE or ACCESS_ONCE() in different C statements.
+ *
+ * In contrast to ACCESS_ONCE these two macros will also work on aggregate
+ * data types like structs or unions. If the size of the accessed data
+ * type exceeds the word size of the machine (e.g., 32 bits or 64 bits)
+ * READ_ONCE() and ASSIGN_ONCE()  will fall back to memcpy and print a
+ * compile-time warning.
+ *
+ * Their two major use cases are: (1) Mediating communication between
+ * process-level code and irq/NMI handlers, all running on the same CPU,
+ * and (2) Ensuring that the compiler does not  fold, spindle, or otherwise
+ * mutilate accesses that either do not require ordering or that interact
+ * with an explicit memory barrier or atomic instruction that provides the
+ * required ordering.
+ */
+
+#define READ_ONCE(x) \
+	({ typeof(x) __val; __read_once_size(&x, &__val, sizeof(__val)); __val; })
+
+#define ASSIGN_ONCE(val, x) \
+	({ typeof(x) __val; __val = val; __assign_once_size(&x, &__val, sizeof(__val)); __val; })
+
 #endif /* __KERNEL__ */
 
 #endif /* __ASSEMBLY__ */
diff --git a/mm/gup.c b/mm/gup.c
index 0ca1df9075ab..a900759cc807 100644
--- a/mm/gup.c
+++ b/mm/gup.c
@@ -968,7 +968,7 @@ static int gup_pud_range(pgd_t pgd, unsigned long addr, unsigned long end,
 
 	pudp = pud_offset(&pgd, addr);
 	do {
-		pud_t pud = ACCESS_ONCE(*pudp);
+		pud_t pud = READ_ONCE(*pudp);
 
 		next = pud_addr_end(addr, end);
 		if (pud_none(pud))
diff --git a/mm/memory.c b/mm/memory.c
index d8aebc52265f..649e7d440bd7 100644
--- a/mm/memory.c
+++ b/mm/memory.c
@@ -3195,7 +3195,16 @@ static int handle_pte_fault(struct mm_struct *mm,
 	pte_t entry;
 	spinlock_t *ptl;
 
-	entry = ACCESS_ONCE(*pte);
+	/*
+	 * some architectures can have larger ptes than wordsize,
+	 * e.g.ppc44x-defconfig has CONFIG_PTE_64BIT=y and CONFIG_32BIT=y,
+	 * so READ_ONCE or ACCESS_ONCE cannot guarantee atomic accesses.
+	 * The code below just needs a consistent view for the ifs and
+	 * we later double check anyway with the ptl lock held. So here
+	 * a barrier will do.
+	 */
+	entry = *pte;
+	barrier();
 	if (!pte_present(entry)) {
 		if (pte_none(entry)) {
 			if (vma->vm_ops) {
diff --git a/mm/rmap.c b/mm/rmap.c
index 45ba250babd8..c5bc241127b2 100644
--- a/mm/rmap.c
+++ b/mm/rmap.c
@@ -583,7 +583,8 @@ pmd_t *mm_find_pmd(struct mm_struct *mm, unsigned long address)
 	 * without holding anon_vma lock for write.  So when looking for a
 	 * genuine pmde (in which to find pte), test present and !THP together.
 	 */
-	pmde = ACCESS_ONCE(*pmd);
+	pmde = *pmd;
+	barrier();
 	if (!pmd_present(pmde) || pmd_trans_huge(pmde))
 		pmd = NULL;
 out: