s390/mm: try to avoid storage key operation in ptep_set_access_flags

The call to pgste_set_key in ptep_set_access_flags can be avoided
if the old pte is found to be valid at the time the new access
rights are set. The function that created the old, valid pte already
completed the required storage key operation.

Signed-off-by: Martin Schwidefsky <[e-mail]>

arch/s390/include/asm/pgtable.h

···
 					pte_t entry, int dirty)
 {
 	pgste_t pgste;
+	pte_t oldpte;
 
-	if (pte_same(*ptep, entry))
+	oldpte = *ptep;
+	if (pte_same(oldpte, entry))
 		return 0;
 	if (mm_has_pgste(vma->vm_mm)) {
 		pgste = pgste_get_lock(ptep);
···
 	ptep_flush_direct(vma->vm_mm, address, ptep);
 
 	if (mm_has_pgste(vma->vm_mm)) {
-		pgste_set_key(ptep, pgste, entry, 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