s390/mm: uninline ptep_xxx functions from pgtable.h

The code in the various ptep_xxx functions has grown quite large,
consolidate them to four out-of-line functions:
  ptep_xchg_direct to exchange a pte with another with immediate flushing
  ptep_xchg_lazy to exchange a pte with another in a batched update
  ptep_modify_prot_start to begin a protection flags update
  ptep_modify_prot_commit to commit a protection flags update
Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>

arch/s390/include/asm/pgtable.h

@@ -680,69 +680,8 @@ static inline void pgste_set(pte_t *ptep, pgste_t pgste)
 #endif
 }
 
-static inline pgste_t pgste_update_all(pte_t pte, pgste_t pgste,
-				       struct mm_struct *mm)
-{
-#ifdef CONFIG_PGSTE
-	unsigned long address, bits, skey;
-
-	if (!mm_use_skey(mm) || pte_val(pte) & _PAGE_INVALID)
-		return pgste;
-	address = pte_val(pte) & PAGE_MASK;
-	skey = (unsigned long) page_get_storage_key(address);
-	bits = skey & (_PAGE_CHANGED | _PAGE_REFERENCED);
-	/* Transfer page changed & referenced bit to guest bits in pgste */
-	pgste_val(pgste) |= bits << 48;		/* GR bit & GC bit */
-	/* Copy page access key and fetch protection bit to pgste */
-	pgste_val(pgste) &= ~(PGSTE_ACC_BITS | PGSTE_FP_BIT);
-	pgste_val(pgste) |= (skey & (_PAGE_ACC_BITS | _PAGE_FP_BIT)) << 56;
-#endif
-	return pgste;
-
-}
-
-static inline void pgste_set_key(pte_t *ptep, pgste_t pgste, pte_t entry,
-				 struct mm_struct *mm)
-{
-#ifdef CONFIG_PGSTE
-	unsigned long address;
-	unsigned long nkey;
-
-	if (!mm_use_skey(mm) || pte_val(entry) & _PAGE_INVALID)
-		return;
-	VM_BUG_ON(!(pte_val(*ptep) & _PAGE_INVALID));
-	address = pte_val(entry) & PAGE_MASK;
-	/*
-	 * Set page access key and fetch protection bit from pgste.
-	 * The guest C/R information is still in the PGSTE, set real
-	 * key C/R to 0.
-	 */
-	nkey = (pgste_val(pgste) & (PGSTE_ACC_BITS | PGSTE_FP_BIT)) >> 56;
-	nkey |= (pgste_val(pgste) & (PGSTE_GR_BIT | PGSTE_GC_BIT)) >> 48;
-	page_set_storage_key(address, nkey, 0);
-#endif
-}
-
-static inline pgste_t pgste_set_pte(pte_t *ptep, pgste_t pgste, pte_t entry)
-{
-	if ((pte_val(entry) & _PAGE_PRESENT) &&
-	    (pte_val(entry) & _PAGE_WRITE) &&
-	    !(pte_val(entry) & _PAGE_INVALID)) {
-		if (!MACHINE_HAS_ESOP) {
-			/*
-			 * Without enhanced suppression-on-protection force
-			 * the dirty bit on for all writable ptes.
-			 */
-			pte_val(entry) |= _PAGE_DIRTY;
-			pte_val(entry) &= ~_PAGE_PROTECT;
-		}
-		if (!(pte_val(entry) & _PAGE_PROTECT))
-			/* This pte allows write access, set user-dirty */
-			pgste_val(pgste) |= PGSTE_UC_BIT;
-	}
-	*ptep = entry;
-	return pgste;
-}
+bool pgste_test_and_clear_dirty(struct mm_struct *, unsigned long address);
+void ptep_ipte_notify(struct mm_struct *, unsigned long addr, pte_t *);
 
 /**
  * struct gmap_struct - guest address space
@@ -791,47 +730,11 @@ int __gmap_link(struct gmap *gmap, unsigned long gaddr, unsigned long vmaddr);
 int gmap_fault(struct gmap *, unsigned long gaddr, unsigned int fault_flags);
 void gmap_discard(struct gmap *, unsigned long from, unsigned long to);
 void __gmap_zap(struct gmap *, unsigned long gaddr);
-bool gmap_test_and_clear_dirty(unsigned long address, struct gmap *);
 
 
 void gmap_register_ipte_notifier(struct gmap_notifier *);
 void gmap_unregister_ipte_notifier(struct gmap_notifier *);
 int gmap_ipte_notify(struct gmap *, unsigned long start, unsigned long len);
-void gmap_do_ipte_notify(struct mm_struct *, unsigned long addr, pte_t *);
-
-static inline pgste_t pgste_ipte_notify(struct mm_struct *mm,
-					unsigned long addr,
-					pte_t *ptep, pgste_t pgste)
-{
-#ifdef CONFIG_PGSTE
-	if (pgste_val(pgste) & PGSTE_IN_BIT) {
-		pgste_val(pgste) &= ~PGSTE_IN_BIT;
-		gmap_do_ipte_notify(mm, addr, ptep);
-	}
-#endif
-	return pgste;
-}
-
-/*
- * Certain architectures need to do special things when PTEs
- * within a page table are directly modified.  Thus, the following
- * hook is made available.
- */
-static inline void set_pte_at(struct mm_struct *mm, unsigned long addr,
-			      pte_t *ptep, pte_t entry)
-{
-	pgste_t pgste;
-
-	if (mm_has_pgste(mm)) {
-		pgste = pgste_get_lock(ptep);
-		pgste_val(pgste) &= ~_PGSTE_GPS_ZERO;
-		pgste_set_key(ptep, pgste, entry, mm);
-		pgste = pgste_set_pte(ptep, pgste, entry);
-		pgste_set_unlock(ptep, pgste);
-	} else {
-		*ptep = entry;
-	}
-}
 
 /*
  * query functions pte_write/pte_dirty/pte_young only work if
@@ -998,95 +901,30 @@ static inline void __ptep_ipte_range(unsigned long address, int nr, pte_t *ptep)
 	} while (nr != 255);
 }
 
-static inline void ptep_flush_direct(struct mm_struct *mm,
-				     unsigned long address, pte_t *ptep)
-{
-	int active, count;
-
-	if (pte_val(*ptep) & _PAGE_INVALID)
-		return;
-	active = (mm == current->active_mm) ? 1 : 0;
-	count = atomic_add_return(0x10000, &mm->context.attach_count);
-	if (MACHINE_HAS_TLB_LC && (count & 0xffff) <= active &&
-	    cpumask_equal(mm_cpumask(mm), cpumask_of(smp_processor_id())))
-		__ptep_ipte_local(address, ptep);
-	else
-		__ptep_ipte(address, ptep);
-	atomic_sub(0x10000, &mm->context.attach_count);
-}
-
-static inline void ptep_flush_lazy(struct mm_struct *mm,
-				   unsigned long address, pte_t *ptep)
-{
-	int active, count;
-
-	if (pte_val(*ptep) & _PAGE_INVALID)
-		return;
-	active = (mm == current->active_mm) ? 1 : 0;
-	count = atomic_add_return(0x10000, &mm->context.attach_count);
-	if ((count & 0xffff) <= active) {
-		pte_val(*ptep) |= _PAGE_INVALID;
-		mm->context.flush_mm = 1;
-	} else
-		__ptep_ipte(address, ptep);
-	atomic_sub(0x10000, &mm->context.attach_count);
-}
-
 /*
- * Get (and clear) the user dirty bit for a pte.
+ * This is hard to understand. ptep_get_and_clear and ptep_clear_flush
+ * both clear the TLB for the unmapped pte. The reason is that
+ * ptep_get_and_clear is used in common code (e.g. change_pte_range)
+ * to modify an active pte. The sequence is
+ *   1) ptep_get_and_clear
+ *   2) set_pte_at
+ *   3) flush_tlb_range
+ * On s390 the tlb needs to get flushed with the modification of the pte
+ * if the pte is active. The only way how this can be implemented is to
+ * have ptep_get_and_clear do the tlb flush. In exchange flush_tlb_range
+ * is a nop.
  */
-static inline int ptep_test_and_clear_user_dirty(struct mm_struct *mm,
-						 unsigned long addr,
-						 pte_t *ptep)
-{
-	pgste_t pgste;
-	pte_t pte;
-	int dirty;
-
-	if (!mm_has_pgste(mm))
-		return 0;
-	pgste = pgste_get_lock(ptep);
-	dirty = !!(pgste_val(pgste) & PGSTE_UC_BIT);
-	pgste_val(pgste) &= ~PGSTE_UC_BIT;
-	pte = *ptep;
-	if (dirty && (pte_val(pte) & _PAGE_PRESENT)) {
-		pgste = pgste_ipte_notify(mm, addr, ptep, pgste);
-		__ptep_ipte(addr, ptep);
-		if (MACHINE_HAS_ESOP || !(pte_val(pte) & _PAGE_WRITE))
-			pte_val(pte) |= _PAGE_PROTECT;
-		else
-			pte_val(pte) |= _PAGE_INVALID;
-		*ptep = pte;
-	}
-	pgste_set_unlock(ptep, pgste);
-	return dirty;
-}
+pte_t ptep_xchg_direct(struct mm_struct *, unsigned long, pte_t *, pte_t);
+pte_t ptep_xchg_lazy(struct mm_struct *, unsigned long, pte_t *, pte_t);
 
 #define __HAVE_ARCH_PTEP_TEST_AND_CLEAR_YOUNG
 static inline int ptep_test_and_clear_young(struct vm_area_struct *vma,
 					    unsigned long addr, pte_t *ptep)
 {
-	pgste_t pgste;
-	pte_t pte;
-	int young;
-
-	if (mm_has_pgste(vma->vm_mm)) {
-		pgste = pgste_get_lock(ptep);
-		pgste = pgste_ipte_notify(vma->vm_mm, addr, ptep, pgste);
-	}
-
-	pte = *ptep;
-	ptep_flush_direct(vma->vm_mm, addr, ptep);
-	young = pte_young(pte);
-
-	if (mm_has_pgste(vma->vm_mm)) {
-		pgste = pgste_update_all(pte, pgste, vma->vm_mm);
-		pgste = pgste_set_pte(ptep, pgste, pte_mkold(pte));
-		pgste_set_unlock(ptep, pgste);
-	} else
-		*ptep = pte_mkold(pte);
+	pte_t pte = *ptep;
 
-	return young;
+	pte = ptep_xchg_direct(vma->vm_mm, addr, ptep, pte_mkold(pte));
+	return pte_young(pte);
 }
 
 #define __HAVE_ARCH_PTEP_CLEAR_YOUNG_FLUSH
@@ -1096,104 +934,22 @@ static inline int ptep_clear_flush_young(struct vm_area_struct *vma,
 	return ptep_test_and_clear_young(vma, address, ptep);
 }
 
-/*
- * This is hard to understand. ptep_get_and_clear and ptep_clear_flush
- * both clear the TLB for the unmapped pte. The reason is that
- * ptep_get_and_clear is used in common code (e.g. change_pte_range)
- * to modify an active pte. The sequence is
- *   1) ptep_get_and_clear
- *   2) set_pte_at
- *   3) flush_tlb_range
- * On s390 the tlb needs to get flushed with the modification of the pte
- * if the pte is active. The only way how this can be implemented is to
- * have ptep_get_and_clear do the tlb flush. In exchange flush_tlb_range
- * is a nop.
- */
 #define __HAVE_ARCH_PTEP_GET_AND_CLEAR
 static inline pte_t ptep_get_and_clear(struct mm_struct *mm,
-				       unsigned long address, pte_t *ptep)
+				       unsigned long addr, pte_t *ptep)
 {
-	pgste_t pgste;
-	pte_t pte;
-
-	if (mm_has_pgste(mm)) {
-		pgste = pgste_get_lock(ptep);
-		pgste = pgste_ipte_notify(mm, address, ptep, pgste);
-	}
-
-	pte = *ptep;
-	ptep_flush_lazy(mm, address, ptep);
-	pte_val(*ptep) = _PAGE_INVALID;
-
-	if (mm_has_pgste(mm)) {
-		pgste = pgste_update_all(pte, pgste, mm);
-		pgste_set_unlock(ptep, pgste);
-	}
-	return pte;
+	return ptep_xchg_lazy(mm, addr, ptep, __pte(_PAGE_INVALID));
 }
 
 #define __HAVE_ARCH_PTEP_MODIFY_PROT_TRANSACTION
-static inline pte_t ptep_modify_prot_start(struct mm_struct *mm,
-					   unsigned long address,
-					   pte_t *ptep)
-{
-	pgste_t pgste;
-	pte_t pte;
-
-	if (mm_has_pgste(mm)) {
-		pgste = pgste_get_lock(ptep);
-		pgste_ipte_notify(mm, address, ptep, pgste);
-	}
-
-	pte = *ptep;
-	ptep_flush_lazy(mm, address, ptep);
-
-	if (mm_has_pgste(mm)) {
-		pgste = pgste_update_all(pte, pgste, mm);
-		pgste_set(ptep, pgste);
-	}
-	return pte;
-}
-
-static inline void ptep_modify_prot_commit(struct mm_struct *mm,
-					   unsigned long address,
-					   pte_t *ptep, pte_t pte)
-{
-	pgste_t pgste;
-
-	if (mm_has_pgste(mm)) {
-		pgste = pgste_get(ptep);
-		pgste_set_key(ptep, pgste, pte, mm);
-		pgste = pgste_set_pte(ptep, pgste, pte);
-		pgste_set_unlock(ptep, pgste);
-	} else
-		*ptep = pte;
-}
+pte_t ptep_modify_prot_start(struct mm_struct *, unsigned long, pte_t *);
+void ptep_modify_prot_commit(struct mm_struct *, unsigned long, pte_t *, pte_t);
 
 #define __HAVE_ARCH_PTEP_CLEAR_FLUSH
 static inline pte_t ptep_clear_flush(struct vm_area_struct *vma,
-				     unsigned long address, pte_t *ptep)
+				     unsigned long addr, pte_t *ptep)
 {
-	pgste_t pgste;
-	pte_t pte;
-
-	if (mm_has_pgste(vma->vm_mm)) {
-		pgste = pgste_get_lock(ptep);
-		pgste = pgste_ipte_notify(vma->vm_mm, address, ptep, pgste);
-	}
-
-	pte = *ptep;
-	ptep_flush_direct(vma->vm_mm, address, ptep);
-	pte_val(*ptep) = _PAGE_INVALID;
-
-	if (mm_has_pgste(vma->vm_mm)) {
-		if ((pgste_val(pgste) & _PGSTE_GPS_USAGE_MASK) ==
-		    _PGSTE_GPS_USAGE_UNUSED)
-			pte_val(pte) |= _PAGE_UNUSED;
-		pgste = pgste_update_all(pte, pgste, vma->vm_mm);
-		pgste_set_unlock(ptep, pgste);
-	}
-	return pte;
+	return ptep_xchg_direct(vma->vm_mm, addr, ptep, __pte(_PAGE_INVALID));
 }
 
 /*
@@ -1205,80 +961,52 @@ static inline pte_t ptep_clear_flush(struct vm_area_struct *vma,
  */
 #define __HAVE_ARCH_PTEP_GET_AND_CLEAR_FULL
 static inline pte_t ptep_get_and_clear_full(struct mm_struct *mm,
-					    unsigned long address,
+					    unsigned long addr,
 					    pte_t *ptep, int full)
 {
-	pgste_t pgste;
-	pte_t pte;
-
-	if (!full && mm_has_pgste(mm)) {
-		pgste = pgste_get_lock(ptep);
-		pgste = pgste_ipte_notify(mm, address, ptep, pgste);
-	}
-
-	pte = *ptep;
-	if (!full)
-		ptep_flush_lazy(mm, address, ptep);
-	pte_val(*ptep) = _PAGE_INVALID;
-
-	if (!full && mm_has_pgste(mm)) {
-		pgste = pgste_update_all(pte, pgste, mm);
-		pgste_set_unlock(ptep, pgste);
+	if (full) {
+		pte_t pte = *ptep;
+		*ptep = __pte(_PAGE_INVALID);
+		return pte;
 	}
-	return pte;
+	return ptep_xchg_lazy(mm, addr, ptep, __pte(_PAGE_INVALID));
 }
 
 #define __HAVE_ARCH_PTEP_SET_WRPROTECT
-static inline pte_t ptep_set_wrprotect(struct mm_struct *mm,
-				       unsigned long address, pte_t *ptep)
+static inline void ptep_set_wrprotect(struct mm_struct *mm,
+				      unsigned long addr, pte_t *ptep)
 {
-	pgste_t pgste;
 	pte_t pte = *ptep;
 
-	if (pte_write(pte)) {
-		if (mm_has_pgste(mm)) {
-			pgste = pgste_get_lock(ptep);
-			pgste = pgste_ipte_notify(mm, address, ptep, pgste);
-		}
-
-		ptep_flush_lazy(mm, address, ptep);
-		pte = pte_wrprotect(pte);
-
-		if (mm_has_pgste(mm)) {
-			pgste = pgste_set_pte(ptep, pgste, pte);
-			pgste_set_unlock(ptep, pgste);
-		} else
-			*ptep = pte;
-	}
-	return pte;
+	if (pte_write(pte))
+		ptep_xchg_lazy(mm, addr, ptep, pte_wrprotect(pte));
 }
 
 #define __HAVE_ARCH_PTEP_SET_ACCESS_FLAGS
 static inline int ptep_set_access_flags(struct vm_area_struct *vma,
-					unsigned long address, pte_t *ptep,
+					unsigned long addr, pte_t *ptep,
 					pte_t entry, int dirty)
 {
-	pgste_t pgste;
-	pte_t oldpte;
-
-	oldpte = *ptep;
-	if (pte_same(oldpte, entry))
+	if (pte_same(*ptep, entry))
 		return 0;
-	if (mm_has_pgste(vma->vm_mm)) {
-		pgste = pgste_get_lock(ptep);
-		pgste = pgste_ipte_notify(vma->vm_mm, address, ptep, pgste);
-	}
+	ptep_xchg_direct(vma->vm_mm, addr, ptep, entry);
+	return 1;
+}
 
-	ptep_flush_direct(vma->vm_mm, address, ptep);
+void set_pte_pgste_at(struct mm_struct *, unsigned long, pte_t *, pte_t);
 
-	if (mm_has_pgste(vma->vm_mm)) {
-		if (pte_val(oldpte) & _PAGE_INVALID)
-			pgste_set_key(ptep, pgste, entry, vma->vm_mm);
-		pgste = pgste_set_pte(ptep, pgste, entry);
-		pgste_set_unlock(ptep, pgste);
-	} else
+/*
+ * Certain architectures need to do special things when PTEs
+ * within a page table are directly modified.  Thus, the following
+ * hook is made available.
+ */
+static inline void set_pte_at(struct mm_struct *mm, unsigned long addr,
+			      pte_t *ptep, pte_t entry)
+{
+	if (mm_has_pgste(mm))
+		set_pte_pgste_at(mm, addr, ptep, entry);
+	else
 		*ptep = entry;
-	return 1;
 }
 
 /*

arch/s390/kvm/kvm-s390.c

@@ -280,7 +280,7 @@ static void kvm_s390_sync_dirty_log(struct kvm *kvm,
 	for (cur_gfn = memslot->base_gfn; cur_gfn <= last_gfn; cur_gfn++) {
 		address = gfn_to_hva_memslot(memslot, cur_gfn);
 
-		if (gmap_test_and_clear_dirty(address, gmap))
+		if (pgste_test_and_clear_dirty(gmap->mm, address))
 			mark_page_dirty(kvm, cur_gfn);
 	}
 	up_read(&gmap->mm->mmap_sem);

arch/s390/mm/pgtable.c

@@ -772,7 +772,7 @@ int gmap_ipte_notify(struct gmap *gmap, unsigned long gaddr, unsigned long len)
 EXPORT_SYMBOL_GPL(gmap_ipte_notify);
 
 /**
- * gmap_do_ipte_notify - call all invalidation callbacks for a specific pte.
+ * ptep_ipte_notify - call all invalidation callbacks for a specific pte.
  * @mm: pointer to the process mm_struct
  * @addr: virtual address in the process address space
  * @pte: pointer to the page table entry
@@ -780,7 +780,7 @@ EXPORT_SYMBOL_GPL(gmap_ipte_notify);
  * This function is assumed to be called with the page table lock held
  * for the pte to notify.
  */
-void gmap_do_ipte_notify(struct mm_struct *mm, unsigned long vmaddr, pte_t *pte)
+void ptep_ipte_notify(struct mm_struct *mm, unsigned long vmaddr, pte_t *pte)
 {
 	unsigned long offset, gaddr;
 	unsigned long *table;
@@ -801,7 +801,7 @@ void gmap_do_ipte_notify(struct mm_struct *mm, unsigned long vmaddr, pte_t *pte)
 	}
 	spin_unlock(&gmap_notifier_lock);
 }
-EXPORT_SYMBOL_GPL(gmap_do_ipte_notify);
+EXPORT_SYMBOL_GPL(ptep_ipte_notify);
 
 int set_guest_storage_key(struct mm_struct *mm, unsigned long addr,
 			  unsigned long key, bool nq)
@@ -1158,6 +1158,266 @@ static inline void thp_split_mm(struct mm_struct *mm)
 }
 #endif /* CONFIG_TRANSPARENT_HUGEPAGE */
 
+static inline pte_t ptep_flush_direct(struct mm_struct *mm,
+				      unsigned long addr, pte_t *ptep)
+{
+	int active, count;
+	pte_t old;
+
+	old = *ptep;
+	if (unlikely(pte_val(old) & _PAGE_INVALID))
+		return old;
+	active = (mm == current->active_mm) ? 1 : 0;
+	count = atomic_add_return(0x10000, &mm->context.attach_count);
+	if (MACHINE_HAS_TLB_LC && (count & 0xffff) <= active &&
+	    cpumask_equal(mm_cpumask(mm), cpumask_of(smp_processor_id())))
+		__ptep_ipte_local(addr, ptep);
+	else
+		__ptep_ipte(addr, ptep);
+	atomic_sub(0x10000, &mm->context.attach_count);
+	return old;
+}
+
+static inline pte_t ptep_flush_lazy(struct mm_struct *mm,
+				    unsigned long addr, pte_t *ptep)
+{
+	int active, count;
+	pte_t old;
+
+	old = *ptep;
+	if (unlikely(pte_val(old) & _PAGE_INVALID))
+		return old;
+	active = (mm == current->active_mm) ? 1 : 0;
+	count = atomic_add_return(0x10000, &mm->context.attach_count);
+	if ((count & 0xffff) <= active) {
+		pte_val(*ptep) |= _PAGE_INVALID;
+		mm->context.flush_mm = 1;
+	} else
+		__ptep_ipte(addr, ptep);
+	atomic_sub(0x10000, &mm->context.attach_count);
+	return old;
+}
+
+static inline pgste_t pgste_update_all(pte_t pte, pgste_t pgste,
+				       struct mm_struct *mm)
+{
+#ifdef CONFIG_PGSTE
+	unsigned long address, bits, skey;
+
+	if (!mm_use_skey(mm) || pte_val(pte) & _PAGE_INVALID)
+		return pgste;
+	address = pte_val(pte) & PAGE_MASK;
+	skey = (unsigned long) page_get_storage_key(address);
+	bits = skey & (_PAGE_CHANGED | _PAGE_REFERENCED);
+	/* Transfer page changed & referenced bit to guest bits in pgste */
+	pgste_val(pgste) |= bits << 48;		/* GR bit & GC bit */
+	/* Copy page access key and fetch protection bit to pgste */
+	pgste_val(pgste) &= ~(PGSTE_ACC_BITS | PGSTE_FP_BIT);
+	pgste_val(pgste) |= (skey & (_PAGE_ACC_BITS | _PAGE_FP_BIT)) << 56;
+#endif
+	return pgste;
+
+}
+
+static inline void pgste_set_key(pte_t *ptep, pgste_t pgste, pte_t entry,
+				 struct mm_struct *mm)
+{
+#ifdef CONFIG_PGSTE
+	unsigned long address;
+	unsigned long nkey;
+
+	if (!mm_use_skey(mm) || pte_val(entry) & _PAGE_INVALID)
+		return;
+	VM_BUG_ON(!(pte_val(*ptep) & _PAGE_INVALID));
+	address = pte_val(entry) & PAGE_MASK;
+	/*
+	 * Set page access key and fetch protection bit from pgste.
+	 * The guest C/R information is still in the PGSTE, set real
+	 * key C/R to 0.
+	 */
+	nkey = (pgste_val(pgste) & (PGSTE_ACC_BITS | PGSTE_FP_BIT)) >> 56;
+	nkey |= (pgste_val(pgste) & (PGSTE_GR_BIT | PGSTE_GC_BIT)) >> 48;
+	page_set_storage_key(address, nkey, 0);
+#endif
+}
+
+static inline pgste_t pgste_set_pte(pte_t *ptep, pgste_t pgste, pte_t entry)
+{
+#ifdef CONFIG_PGSTE
+	if ((pte_val(entry) & _PAGE_PRESENT) &&
+	    (pte_val(entry) & _PAGE_WRITE) &&
+	    !(pte_val(entry) & _PAGE_INVALID)) {
+		if (!MACHINE_HAS_ESOP) {
+			/*
+			 * Without enhanced suppression-on-protection force
+			 * the dirty bit on for all writable ptes.
+			 */
+			pte_val(entry) |= _PAGE_DIRTY;
+			pte_val(entry) &= ~_PAGE_PROTECT;
+		}
+		if (!(pte_val(entry) & _PAGE_PROTECT))
+			/* This pte allows write access, set user-dirty */
+			pgste_val(pgste) |= PGSTE_UC_BIT;
+	}
+#endif
+	*ptep = entry;
+	return pgste;
+}
+
+static inline pgste_t pgste_ipte_notify(struct mm_struct *mm,
+					unsigned long addr,
+					pte_t *ptep, pgste_t pgste)
+{
+#ifdef CONFIG_PGSTE
+	if (pgste_val(pgste) & PGSTE_IN_BIT) {
+		pgste_val(pgste) &= ~PGSTE_IN_BIT;
+		ptep_ipte_notify(mm, addr, ptep);
+	}
+#endif
+	return pgste;
+}
+
+#ifdef CONFIG_PGSTE
+/*
+ * Test and reset if a guest page is dirty
+ */
+bool pgste_test_and_clear_dirty(struct mm_struct *mm, unsigned long addr)
+{
+	spinlock_t *ptl;
+	pgste_t pgste;
+	pte_t *ptep;
+	pte_t pte;
+	bool dirty;
+
+	ptep = get_locked_pte(mm, addr, &ptl);
+	if (unlikely(!ptep))
+		return false;
+
+	pgste = pgste_get_lock(ptep);
+	dirty = !!(pgste_val(pgste) & PGSTE_UC_BIT);
+	pgste_val(pgste) &= ~PGSTE_UC_BIT;
+	pte = *ptep;
+	if (dirty && (pte_val(pte) & _PAGE_PRESENT)) {
+		pgste = pgste_ipte_notify(mm, addr, ptep, pgste);
+		__ptep_ipte(addr, ptep);
+		if (MACHINE_HAS_ESOP || !(pte_val(pte) & _PAGE_WRITE))
+			pte_val(pte) |= _PAGE_PROTECT;
+		else
+			pte_val(pte) |= _PAGE_INVALID;
+		*ptep = pte;
+	}
+	pgste_set_unlock(ptep, pgste);
+
+	spin_unlock(ptl);
+	return dirty;
+}
+EXPORT_SYMBOL_GPL(pgste_test_and_clear_dirty);
+
+void set_pte_pgste_at(struct mm_struct *mm, unsigned long addr,
+		      pte_t *ptep, pte_t entry)
+{
+	pgste_t pgste;
+
+	/* the mm_has_pgste() check is done in set_pte_at() */
+	pgste = pgste_get_lock(ptep);
+	pgste_val(pgste) &= ~_PGSTE_GPS_ZERO;
+	pgste_set_key(ptep, pgste, entry, mm);
+	pgste = pgste_set_pte(ptep, pgste, entry);
+	pgste_set_unlock(ptep, pgste);
+}
+EXPORT_SYMBOL(set_pte_pgste_at);
+#endif
+
+static inline pgste_t ptep_xchg_start(struct mm_struct *mm,
+				      unsigned long addr, pte_t *ptep)
+{
+	pgste_t pgste = __pgste(0);
+
+	if (mm_has_pgste(mm)) {
+		pgste = pgste_get_lock(ptep);
+		pgste = pgste_ipte_notify(mm, addr, ptep, pgste);
+	}
+	return pgste;
+}
+
+static inline void ptep_xchg_commit(struct mm_struct *mm,
+				    unsigned long addr, pte_t *ptep,
+				    pgste_t pgste, pte_t old, pte_t new)
+{
+	if (mm_has_pgste(mm)) {
+		if (pte_val(old) & _PAGE_INVALID)
+			pgste_set_key(ptep, pgste, new, mm);
+		if (pte_val(new) & _PAGE_INVALID) {
+			pgste = pgste_update_all(old, pgste, mm);
+			if ((pgste_val(pgste) & _PGSTE_GPS_USAGE_MASK) ==
+			    _PGSTE_GPS_USAGE_UNUSED)
+				pte_val(old) |= _PAGE_UNUSED;
+		}
+		pgste = pgste_set_pte(ptep, pgste, new);
+		pgste_set_unlock(ptep, pgste);
+	} else {
+		*ptep = new;
+	}
+}
+
+pte_t ptep_xchg_direct(struct mm_struct *mm, unsigned long addr,
+		       pte_t *ptep, pte_t new)
+{
+	pgste_t pgste;
+	pte_t old;
+
+	pgste = ptep_xchg_start(mm, addr, ptep);
+	old = ptep_flush_direct(mm, addr, ptep);
+	ptep_xchg_commit(mm, addr, ptep, pgste, old, new);
+	return old;
+}
+EXPORT_SYMBOL(ptep_xchg_direct);
+
+pte_t ptep_xchg_lazy(struct mm_struct *mm, unsigned long addr,
+		     pte_t *ptep, pte_t new)
+{
+	pgste_t pgste;
+	pte_t old;
+
+	pgste = ptep_xchg_start(mm, addr, ptep);
+	old = ptep_flush_lazy(mm, addr, ptep);
+	ptep_xchg_commit(mm, addr, ptep, pgste, old, new);
+	return old;
+}
+EXPORT_SYMBOL(ptep_xchg_lazy);
+
+pte_t ptep_modify_prot_start(struct mm_struct *mm, unsigned long addr,
+			     pte_t *ptep)
+{
+	pgste_t pgste;
+	pte_t old;
+
+	pgste = ptep_xchg_start(mm, addr, ptep);
+	old = ptep_flush_lazy(mm, addr, ptep);
+	if (mm_has_pgste(mm)) {
+		pgste = pgste_update_all(old, pgste, mm);
+		pgste_set(ptep, pgste);
+	}
+	return old;
+}
+EXPORT_SYMBOL(ptep_modify_prot_start);
+
+void ptep_modify_prot_commit(struct mm_struct *mm, unsigned long addr,
+			     pte_t *ptep, pte_t pte)
+{
+	pgste_t pgste;
+
+	if (mm_has_pgste(mm)) {
+		pgste = pgste_get(ptep);
+		pgste_set_key(ptep, pgste, pte, mm);
+		pgste = pgste_set_pte(ptep, pgste, pte);
+		pgste_set_unlock(ptep, pgste);
+	} else {
+		*ptep = pte;
+	}
+}
+EXPORT_SYMBOL(ptep_modify_prot_commit);
+
 /*
  * switch on pgstes for its userspace process (for kvm)
  */
@@ -1190,17 +1450,15 @@ static int __s390_enable_skey(pte_t *pte, unsigned long addr,
 	unsigned long ptev;
 	pgste_t pgste;
 
-	pgste = pgste_get_lock(pte);
 	/*
 	 * Remove all zero page mappings,
 	 * after establishing a policy to forbid zero page mappings
 	 * following faults for that page will get fresh anonymous pages
 	 */
-	if (is_zero_pfn(pte_pfn(*pte))) {
-		ptep_flush_direct(walk->mm, addr, pte);
-		pte_val(*pte) = _PAGE_INVALID;
-	}
+	if (is_zero_pfn(pte_pfn(*pte)))
+		ptep_xchg_direct(walk->mm, addr, pte, __pte(_PAGE_INVALID));
 	/* Clear storage key */
+	pgste = pgste_get_lock(pte);
 	pgste_val(pgste) &= ~(PGSTE_ACC_BITS | PGSTE_FP_BIT |
 			      PGSTE_GR_BIT | PGSTE_GC_BIT);
 	ptev = pte_val(*pte);
@@ -1266,27 +1524,6 @@ void s390_reset_cmma(struct mm_struct *mm)
 }
 EXPORT_SYMBOL_GPL(s390_reset_cmma);
 
-/*
- * Test and reset if a guest page is dirty
- */
-bool gmap_test_and_clear_dirty(unsigned long address, struct gmap *gmap)
-{
-	pte_t *pte;
-	spinlock_t *ptl;
-	bool dirty = false;
-
-	pte = get_locked_pte(gmap->mm, address, &ptl);
-	if (unlikely(!pte))
-		return false;
-
-	if (ptep_test_and_clear_user_dirty(gmap->mm, address, pte))
-		dirty = true;
-
-	spin_unlock(ptl);
-	return dirty;
-}
-EXPORT_SYMBOL_GPL(gmap_test_and_clear_dirty);
-
 #ifdef CONFIG_TRANSPARENT_HUGEPAGE
 int pmdp_clear_flush_young(struct vm_area_struct *vma, unsigned long address,
 			   pmd_t *pmdp)