s390/mm: cleanup page table definitions

Improve the encoding of the different pte types and the naming of the
page, segment table and region table bits. Due to the different pte
encoding the hugetlbfs primitives need to be adapted as well. To improve
compatability with common code make the huge ptes use the encoding of
normal ptes. The conversion between the pte and pmd encoding for a huge
pte is done with set_huge_pte_at and huge_ptep_get.
Overall the code is now easier to understand.

Reviewed-by: Gerald Schaefer <gerald.schaefer@de.ibm.com>
Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>

arch/s390/include/asm/hugetlb.h

@@ -17,6 +17,9 @@
 
 void set_huge_pte_at(struct mm_struct *mm, unsigned long addr,
 		     pte_t *ptep, pte_t pte);
+pte_t huge_ptep_get(pte_t *ptep);
+pte_t huge_ptep_get_and_clear(struct mm_struct *mm,
+			      unsigned long addr, pte_t *ptep);
 
 /*
  * If the arch doesn't supply something else, assume that hugepage
@@ -38,147 +41,75 @@ static inline int prepare_hugepage_range(struct file *file,
 int arch_prepare_hugepage(struct page *page);
 void arch_release_hugepage(struct page *page);
 
-static inline pte_t huge_pte_wrprotect(pte_t pte)
+static inline void huge_pte_clear(struct mm_struct *mm, unsigned long addr,
+				  pte_t *ptep)
 {
-	pte_val(pte) |= _PAGE_RO;
-	return pte;
+	pte_val(*ptep) = _SEGMENT_ENTRY_EMPTY;
 }
 
-static inline int huge_pte_none(pte_t pte)
+static inline void huge_ptep_clear_flush(struct vm_area_struct *vma,
+					 unsigned long address, pte_t *ptep)
 {
-	return (pte_val(pte) & _SEGMENT_ENTRY_INV) &&
-		!(pte_val(pte) & _SEGMENT_ENTRY_RO);
+	huge_ptep_get_and_clear(vma->vm_mm, address, ptep);
 }
 
-static inline pte_t huge_ptep_get(pte_t *ptep)
+static inline int huge_ptep_set_access_flags(struct vm_area_struct *vma,
+					     unsigned long addr, pte_t *ptep,
+					     pte_t pte, int dirty)
 {
-	pte_t pte = *ptep;
-	unsigned long mask;
-
-	if (!MACHINE_HAS_HPAGE) {
-		ptep = (pte_t *) (pte_val(pte) & _SEGMENT_ENTRY_ORIGIN);
-		if (ptep) {
-			mask = pte_val(pte) &
-				(_SEGMENT_ENTRY_INV | _SEGMENT_ENTRY_RO);
-			pte = pte_mkhuge(*ptep);
-			pte_val(pte) |= mask;
-		}
+	int changed = !pte_same(huge_ptep_get(ptep), pte);
+	if (changed) {
+		huge_ptep_get_and_clear(vma->vm_mm, addr, ptep);
+		set_huge_pte_at(vma->vm_mm, addr, ptep, pte);
 	}
-	return pte;
+	return changed;
 }
 
-static inline void __pmd_csp(pmd_t *pmdp)
+static inline void huge_ptep_set_wrprotect(struct mm_struct *mm,
+					   unsigned long addr, pte_t *ptep)
 {
-	register unsigned long reg2 asm("2") = pmd_val(*pmdp);
-	register unsigned long reg3 asm("3") = pmd_val(*pmdp) |
-					       _SEGMENT_ENTRY_INV;
-	register unsigned long reg4 asm("4") = ((unsigned long) pmdp) + 5;
-
-	asm volatile(
-		"	csp %1,%3"
-		: "=m" (*pmdp)
-		: "d" (reg2), "d" (reg3), "d" (reg4), "m" (*pmdp) : "cc");
+	pte_t pte = huge_ptep_get_and_clear(mm, addr, ptep);
+	set_huge_pte_at(mm, addr, ptep, pte_wrprotect(pte));
 }
 
-static inline void huge_ptep_invalidate(struct mm_struct *mm,
-					unsigned long address, pte_t *ptep)
-{
-	pmd_t *pmdp = (pmd_t *) ptep;
-
-	if (MACHINE_HAS_IDTE)
-		__pmd_idte(address, pmdp);
-	else
-		__pmd_csp(pmdp);
-	pmd_val(*pmdp) = _SEGMENT_ENTRY_INV | _SEGMENT_ENTRY;
-}
-
-static inline pte_t huge_ptep_get_and_clear(struct mm_struct *mm,
-					    unsigned long addr, pte_t *ptep)
-{
-	pte_t pte = huge_ptep_get(ptep);
-
-	huge_ptep_invalidate(mm, addr, ptep);
-	return pte;
-}
-
-#define huge_ptep_set_access_flags(__vma, __addr, __ptep, __entry, __dirty) \
-({									    \
-	int __changed = !pte_same(huge_ptep_get(__ptep), __entry);	    \
-	if (__changed) {						    \
-		huge_ptep_invalidate((__vma)->vm_mm, __addr, __ptep);	    \
-		set_huge_pte_at((__vma)->vm_mm, __addr, __ptep, __entry);   \
-	}								    \
-	__changed;							    \
-})
-
-#define huge_ptep_set_wrprotect(__mm, __addr, __ptep)			\
-({									\
-	pte_t __pte = huge_ptep_get(__ptep);				\
-	if (huge_pte_write(__pte)) {					\
-		huge_ptep_invalidate(__mm, __addr, __ptep);		\
-		set_huge_pte_at(__mm, __addr, __ptep,			\
-				huge_pte_wrprotect(__pte));		\
-	}								\
-})
-
-static inline void huge_ptep_clear_flush(struct vm_area_struct *vma,
-					 unsigned long address, pte_t *ptep)
+static inline pte_t mk_huge_pte(struct page *page, pgprot_t pgprot)
 {
-	huge_ptep_invalidate(vma->vm_mm, address, ptep);
+	return mk_pte(page, pgprot);
 }
 
-static inline pte_t mk_huge_pte(struct page *page, pgprot_t pgprot)
+static inline int huge_pte_none(pte_t pte)
 {
-	pte_t pte;
-	pmd_t pmd;
-
-	pmd = mk_pmd_phys(page_to_phys(page), pgprot);
-	pte_val(pte) = pmd_val(pmd);
-	return pte;
+	return pte_none(pte);
 }
 
 static inline int huge_pte_write(pte_t pte)
 {
-	pmd_t pmd;
-
-	pmd_val(pmd) = pte_val(pte);
-	return pmd_write(pmd);
+	return pte_write(pte);
 }
 
 static inline int huge_pte_dirty(pte_t pte)
 {
-	/* No dirty bit in the segment table entry. */
-	return 0;
+	return pte_dirty(pte);
 }
 
 static inline pte_t huge_pte_mkwrite(pte_t pte)
 {
-	pmd_t pmd;
-
-	pmd_val(pmd) = pte_val(pte);
-	pte_val(pte) = pmd_val(pmd_mkwrite(pmd));
-	return pte;
+	return pte_mkwrite(pte);
 }
 
 static inline pte_t huge_pte_mkdirty(pte_t pte)
 {
-	/* No dirty bit in the segment table entry. */
-	return pte;
+	return pte_mkdirty(pte);
 }
 
-static inline pte_t huge_pte_modify(pte_t pte, pgprot_t newprot)
+static inline pte_t huge_pte_wrprotect(pte_t pte)
 {
-	pmd_t pmd;
-
-	pmd_val(pmd) = pte_val(pte);
-	pte_val(pte) = pmd_val(pmd_modify(pmd, newprot));
-	return pte;
+	return pte_wrprotect(pte);
 }
 
-static inline void huge_pte_clear(struct mm_struct *mm, unsigned long addr,
-				  pte_t *ptep)
+static inline pte_t huge_pte_modify(pte_t pte, pgprot_t newprot)
 {
-	pmd_clear((pmd_t *) ptep);
+	return pte_modify(pte, newprot);
 }
 
 #endif /* _ASM_S390_HUGETLB_H */

arch/s390/include/asm/pgtable.h

@@ -217,63 +217,50 @@ extern unsigned long MODULES_END;
 
 /* Hardware bits in the page table entry */
 #define _PAGE_CO	0x100		/* HW Change-bit override */
-#define _PAGE_RO	0x200		/* HW read-only bit  */
+#define _PAGE_PROTECT	0x200		/* HW read-only bit  */
 #define _PAGE_INVALID	0x400		/* HW invalid bit    */
+#define _PAGE_LARGE	0x800		/* Bit to mark a large pte */
 
 /* Software bits in the page table entry */
-#define _PAGE_SWT	0x001		/* SW pte type bit t */
-#define _PAGE_SWX	0x002		/* SW pte type bit x */
-#define _PAGE_SWC	0x004		/* SW pte changed bit */
-#define _PAGE_SWR	0x008		/* SW pte referenced bit */
-#define _PAGE_SWW	0x010		/* SW pte write bit */
+#define _PAGE_PRESENT	0x001		/* SW pte present bit */
+#define _PAGE_TYPE	0x002		/* SW pte type bit */
+#define _PAGE_YOUNG	0x004		/* SW pte young bit */
+#define _PAGE_DIRTY	0x008		/* SW pte dirty bit */
+#define _PAGE_WRITE	0x010		/* SW pte write bit */
 #define _PAGE_SPECIAL	0x020		/* SW associated with special page */
 #define __HAVE_ARCH_PTE_SPECIAL
 
 /* Set of bits not changed in pte_modify */
 #define _PAGE_CHG_MASK		(PAGE_MASK | _PAGE_SPECIAL | _PAGE_CO | \
-				 _PAGE_SWC | _PAGE_SWR)
-
-/* Six different types of pages. */
-#define _PAGE_TYPE_EMPTY	0x400
-#define _PAGE_TYPE_NONE		0x401
-#define _PAGE_TYPE_SWAP		0x403
-#define _PAGE_TYPE_FILE		0x601	/* bit 0x002 is used for offset !! */
-#define _PAGE_TYPE_RO		0x200
-#define _PAGE_TYPE_RW		0x000
-
-/*
- * Only four types for huge pages, using the invalid bit and protection bit
- * of a segment table entry.
- */
-#define _HPAGE_TYPE_EMPTY	0x020	/* _SEGMENT_ENTRY_INV */
-#define _HPAGE_TYPE_NONE	0x220
-#define _HPAGE_TYPE_RO		0x200	/* _SEGMENT_ENTRY_RO  */
-#define _HPAGE_TYPE_RW		0x000
+				 _PAGE_DIRTY | _PAGE_YOUNG)
 
 /*
- * PTE type bits are rather complicated. handle_pte_fault uses pte_present,
- * pte_none and pte_file to find out the pte type WITHOUT holding the page
- * table lock. ptep_clear_flush on the other hand uses ptep_clear_flush to
- * invalidate a given pte. ipte sets the hw invalid bit and clears all tlbs
- * for the page. The page table entry is set to _PAGE_TYPE_EMPTY afterwards.
- * This change is done while holding the lock, but the intermediate step
- * of a previously valid pte with the hw invalid bit set can be observed by
- * handle_pte_fault. That makes it necessary that all valid pte types with
- * the hw invalid bit set must be distinguishable from the four pte types
- * empty, none, swap and file.
+ * handle_pte_fault uses pte_present, pte_none and pte_file to find out the
+ * pte type WITHOUT holding the page table lock. The _PAGE_PRESENT bit
+ * is used to distinguish present from not-present ptes. It is changed only
+ * with the page table lock held.
  *
- *			irxt  ipte  irxt
- * _PAGE_TYPE_EMPTY	1000   ->   1000
- * _PAGE_TYPE_NONE	1001   ->   1001
- * _PAGE_TYPE_SWAP	1011   ->   1011
- * _PAGE_TYPE_FILE	11?1   ->   11?1
- * _PAGE_TYPE_RO	0100   ->   1100
- * _PAGE_TYPE_RW	0000   ->   1000
+ * The following table gives the different possible bit combinations for
+ * the pte hardware and software bits in the last 12 bits of a pte:
  *
- * pte_none is true for bits combinations 1000, 1010, 1100, 1110
- * pte_present is true for bits combinations 0000, 0010, 0100, 0110, 1001
- * pte_file is true for bits combinations 1101, 1111
- * swap pte is 1011 and 0001, 0011, 0101, 0111 are invalid.
+ *			842100000000
+ *			000084210000
+ *			000000008421
+ *			.IR....wdytp
+ * empty		.10....00000
+ * swap			.10....xxx10
+ * file			.11....xxxx0
+ * prot-none, clean	.11....00x01
+ * prot-none, dirty	.10....01x01
+ * read-only, clean	.01....00x01
+ * read-only, dirty	.01....01x01
+ * read-write, clean	.01....10x01
+ * read-write, dirty	.00....11x01
+ *
+ * pte_present is true for the bit pattern .xx...xxxxx1, (pte & 0x001) == 0x001
+ * pte_none    is true for the bit pattern .10...xxxx00, (pte & 0x603) == 0x400
+ * pte_file    is true for the bit pattern .11...xxxxx0, (pte & 0x601) == 0x600
+ * pte_swap    is true for the bit pattern .10...xxxx10, (pte & 0x603) == 0x402
  */
 
 #ifndef CONFIG_64BIT
@@ -287,13 +274,13 @@ extern unsigned long MODULES_END;
 
 /* Bits in the segment table entry */
 #define _SEGMENT_ENTRY_ORIGIN	0x7fffffc0UL	/* page table origin	    */
-#define _SEGMENT_ENTRY_RO	0x200	/* page protection bit		    */
-#define _SEGMENT_ENTRY_INV	0x20	/* invalid segment table entry	    */
+#define _SEGMENT_ENTRY_PROTECT	0x200	/* page protection bit		    */
+#define _SEGMENT_ENTRY_INVALID	0x20	/* invalid segment table entry	    */
 #define _SEGMENT_ENTRY_COMMON	0x10	/* common segment bit		    */
 #define _SEGMENT_ENTRY_PTL	0x0f	/* page table length		    */
 
 #define _SEGMENT_ENTRY		(_SEGMENT_ENTRY_PTL)
-#define _SEGMENT_ENTRY_EMPTY	(_SEGMENT_ENTRY_INV)
+#define _SEGMENT_ENTRY_EMPTY	(_SEGMENT_ENTRY_INVALID)
 
 /* Page status table bits for virtualization */
 #define PGSTE_ACC_BITS	0xf0000000UL
@@ -324,8 +311,8 @@ extern unsigned long MODULES_END;
 
 /* Bits in the region table entry */
 #define _REGION_ENTRY_ORIGIN	~0xfffUL/* region/segment table origin	    */
-#define _REGION_ENTRY_RO	0x200	/* region protection bit	    */
-#define _REGION_ENTRY_INV	0x20	/* invalid region table entry	    */
+#define _REGION_ENTRY_PROTECT	0x200	/* region protection bit	    */
+#define _REGION_ENTRY_INVALID	0x20	/* invalid region table entry	    */
 #define _REGION_ENTRY_TYPE_MASK	0x0c	/* region/segment table type mask   */
 #define _REGION_ENTRY_TYPE_R1	0x0c	/* region first table type	    */
 #define _REGION_ENTRY_TYPE_R2	0x08	/* region second table type	    */
@@ -333,11 +320,11 @@ extern unsigned long MODULES_END;
 #define _REGION_ENTRY_LENGTH	0x03	/* region third length		    */
 
 #define _REGION1_ENTRY		(_REGION_ENTRY_TYPE_R1 | _REGION_ENTRY_LENGTH)
-#define _REGION1_ENTRY_EMPTY	(_REGION_ENTRY_TYPE_R1 | _REGION_ENTRY_INV)
+#define _REGION1_ENTRY_EMPTY	(_REGION_ENTRY_TYPE_R1 | _REGION_ENTRY_INVALID)
 #define _REGION2_ENTRY		(_REGION_ENTRY_TYPE_R2 | _REGION_ENTRY_LENGTH)
-#define _REGION2_ENTRY_EMPTY	(_REGION_ENTRY_TYPE_R2 | _REGION_ENTRY_INV)
+#define _REGION2_ENTRY_EMPTY	(_REGION_ENTRY_TYPE_R2 | _REGION_ENTRY_INVALID)
 #define _REGION3_ENTRY		(_REGION_ENTRY_TYPE_R3 | _REGION_ENTRY_LENGTH)
-#define _REGION3_ENTRY_EMPTY	(_REGION_ENTRY_TYPE_R3 | _REGION_ENTRY_INV)
+#define _REGION3_ENTRY_EMPTY	(_REGION_ENTRY_TYPE_R3 | _REGION_ENTRY_INVALID)
 
 #define _REGION3_ENTRY_LARGE	0x400	/* RTTE-format control, large page  */
 #define _REGION3_ENTRY_RO	0x200	/* page protection bit		    */
@@ -346,16 +333,17 @@ extern unsigned long MODULES_END;
 /* Bits in the segment table entry */
 #define _SEGMENT_ENTRY_ORIGIN_LARGE ~0xfffffUL /* large page address	    */
 #define _SEGMENT_ENTRY_ORIGIN	~0x7ffUL/* segment table origin		    */
-#define _SEGMENT_ENTRY_RO	0x200	/* page protection bit		    */
-#define _SEGMENT_ENTRY_INV	0x20	/* invalid segment table entry	    */
+#define _SEGMENT_ENTRY_PROTECT	0x200	/* page protection bit		    */
+#define _SEGMENT_ENTRY_INVALID	0x20	/* invalid segment table entry	    */
 
 #define _SEGMENT_ENTRY		(0)
-#define _SEGMENT_ENTRY_EMPTY	(_SEGMENT_ENTRY_INV)
+#define _SEGMENT_ENTRY_EMPTY	(_SEGMENT_ENTRY_INVALID)
 
 #define _SEGMENT_ENTRY_LARGE	0x400	/* STE-format control, large page   */
 #define _SEGMENT_ENTRY_CO	0x100	/* change-recording override   */
+#define _SEGMENT_ENTRY_SPLIT	0x001	/* THP splitting bit */
+
 #define _SEGMENT_ENTRY_SPLIT_BIT 0	/* THP splitting bit number */
-#define _SEGMENT_ENTRY_SPLIT	(1UL << _SEGMENT_ENTRY_SPLIT_BIT)
 
 /* Set of bits not changed in pmd_modify */
 #define _SEGMENT_CHG_MASK	(_SEGMENT_ENTRY_ORIGIN | _SEGMENT_ENTRY_LARGE \
@@ -386,14 +374,13 @@ extern unsigned long MODULES_END;
 /*
  * Page protection definitions.
  */
-#define PAGE_NONE	__pgprot(_PAGE_TYPE_NONE)
-#define PAGE_RO		__pgprot(_PAGE_TYPE_RO)
-#define PAGE_RW		__pgprot(_PAGE_TYPE_RO | _PAGE_SWW)
-#define PAGE_RWC	__pgprot(_PAGE_TYPE_RW | _PAGE_SWW | _PAGE_SWC)
+#define PAGE_NONE	__pgprot(_PAGE_PRESENT | _PAGE_INVALID)
+#define PAGE_READ	__pgprot(_PAGE_PRESENT | _PAGE_PROTECT)
+#define PAGE_WRITE	__pgprot(_PAGE_PRESENT | _PAGE_WRITE | _PAGE_PROTECT)
 
-#define PAGE_KERNEL	PAGE_RWC
-#define PAGE_SHARED	PAGE_KERNEL
-#define PAGE_COPY	PAGE_RO
+#define PAGE_SHARED	__pgprot(_PAGE_PRESENT | _PAGE_WRITE | _PAGE_DIRTY)
+#define PAGE_KERNEL	__pgprot(_PAGE_PRESENT | _PAGE_WRITE | _PAGE_DIRTY)
+#define PAGE_KERNEL_RO	__pgprot(_PAGE_PRESENT | _PAGE_PROTECT)
 
 /*
  * On s390 the page table entry has an invalid bit and a read-only bit.
@@ -402,29 +389,30 @@ extern unsigned long MODULES_END;
  */
          /*xwr*/
 #define __P000	PAGE_NONE
-#define __P001	PAGE_RO
-#define __P010	PAGE_RO
-#define __P011	PAGE_RO
-#define __P100	PAGE_RO
-#define __P101	PAGE_RO
-#define __P110	PAGE_RO
-#define __P111	PAGE_RO
+#define __P001	PAGE_READ
+#define __P010	PAGE_READ
+#define __P011	PAGE_READ
+#define __P100	PAGE_READ
+#define __P101	PAGE_READ
+#define __P110	PAGE_READ
+#define __P111	PAGE_READ
 
 #define __S000	PAGE_NONE
-#define __S001	PAGE_RO
-#define __S010	PAGE_RW
-#define __S011	PAGE_RW
-#define __S100	PAGE_RO
-#define __S101	PAGE_RO
-#define __S110	PAGE_RW
-#define __S111	PAGE_RW
+#define __S001	PAGE_READ
+#define __S010	PAGE_WRITE
+#define __S011	PAGE_WRITE
+#define __S100	PAGE_READ
+#define __S101	PAGE_READ
+#define __S110	PAGE_WRITE
+#define __S111	PAGE_WRITE
 
 /*
  * Segment entry (large page) protection definitions.
  */
-#define SEGMENT_NONE	__pgprot(_HPAGE_TYPE_NONE)
-#define SEGMENT_RO	__pgprot(_HPAGE_TYPE_RO)
-#define SEGMENT_RW	__pgprot(_HPAGE_TYPE_RW)
+#define SEGMENT_NONE	__pgprot(_SEGMENT_ENTRY_INVALID | \
+				 _SEGMENT_ENTRY_PROTECT)
+#define SEGMENT_READ	__pgprot(_SEGMENT_ENTRY_PROTECT)
+#define SEGMENT_WRITE	__pgprot(0)
 
 static inline int mm_exclusive(struct mm_struct *mm)
 {
@@ -467,7 +455,7 @@ static inline int pgd_none(pgd_t pgd)
 {
 	if ((pgd_val(pgd) & _REGION_ENTRY_TYPE_MASK) < _REGION_ENTRY_TYPE_R2)
 		return 0;
-	return (pgd_val(pgd) & _REGION_ENTRY_INV) != 0UL;
+	return (pgd_val(pgd) & _REGION_ENTRY_INVALID) != 0UL;
 }
 
 static inline int pgd_bad(pgd_t pgd)
@@ -478,7 +466,7 @@ static inline int pgd_bad(pgd_t pgd)
 	 * invalid for either table entry.
 	 */
 	unsigned long mask =
-		~_SEGMENT_ENTRY_ORIGIN & ~_REGION_ENTRY_INV &
+		~_SEGMENT_ENTRY_ORIGIN & ~_REGION_ENTRY_INVALID &
 		~_REGION_ENTRY_TYPE_MASK & ~_REGION_ENTRY_LENGTH;
 	return (pgd_val(pgd) & mask) != 0;
 }
@@ -494,7 +482,7 @@ static inline int pud_none(pud_t pud)
 {
 	if ((pud_val(pud) & _REGION_ENTRY_TYPE_MASK) < _REGION_ENTRY_TYPE_R3)
 		return 0;
-	return (pud_val(pud) & _REGION_ENTRY_INV) != 0UL;
+	return (pud_val(pud) & _REGION_ENTRY_INVALID) != 0UL;
 }
 
 static inline int pud_large(pud_t pud)
@@ -512,7 +500,7 @@ static inline int pud_bad(pud_t pud)
 	 * invalid for either table entry.
 	 */
 	unsigned long mask =
-		~_SEGMENT_ENTRY_ORIGIN & ~_REGION_ENTRY_INV &
+		~_SEGMENT_ENTRY_ORIGIN & ~_REGION_ENTRY_INVALID &
 		~_REGION_ENTRY_TYPE_MASK & ~_REGION_ENTRY_LENGTH;
 	return (pud_val(pud) & mask) != 0;
 }
@@ -521,21 +509,18 @@ static inline int pud_bad(pud_t pud)
 
 static inline int pmd_present(pmd_t pmd)
 {
-	unsigned long mask = _SEGMENT_ENTRY_INV | _SEGMENT_ENTRY_RO;
-	return (pmd_val(pmd) & mask) == _HPAGE_TYPE_NONE ||
-	       !(pmd_val(pmd) & _SEGMENT_ENTRY_INV);
+	return pmd_val(pmd) != _SEGMENT_ENTRY_INVALID;
 }
 
 static inline int pmd_none(pmd_t pmd)
 {
-	return (pmd_val(pmd) & _SEGMENT_ENTRY_INV) &&
-	       !(pmd_val(pmd) & _SEGMENT_ENTRY_RO);
+	return pmd_val(pmd) == _SEGMENT_ENTRY_INVALID;
 }
 
 static inline int pmd_large(pmd_t pmd)
 {
 #ifdef CONFIG_64BIT
-	return !!(pmd_val(pmd) & _SEGMENT_ENTRY_LARGE);
+	return (pmd_val(pmd) & _SEGMENT_ENTRY_LARGE) != 0;
 #else
 	return 0;
 #endif
@@ -543,7 +528,7 @@ static inline int pmd_large(pmd_t pmd)
 
 static inline int pmd_bad(pmd_t pmd)
 {
-	unsigned long mask = ~_SEGMENT_ENTRY_ORIGIN & ~_SEGMENT_ENTRY_INV;
+	unsigned long mask = ~_SEGMENT_ENTRY_ORIGIN & ~_SEGMENT_ENTRY_INVALID;
 	return (pmd_val(pmd) & mask) != _SEGMENT_ENTRY;
 }
 
@@ -563,7 +548,7 @@ extern int pmdp_clear_flush_young(struct vm_area_struct *vma,
 #define __HAVE_ARCH_PMD_WRITE
 static inline int pmd_write(pmd_t pmd)
 {
-	return (pmd_val(pmd) & _SEGMENT_ENTRY_RO) == 0;
+	return (pmd_val(pmd) & _SEGMENT_ENTRY_PROTECT) == 0;
 }
 
 static inline int pmd_young(pmd_t pmd)
@@ -571,23 +556,23 @@ static inline int pmd_young(pmd_t pmd)
 	return 0;
 }
 
-static inline int pte_none(pte_t pte)
+static inline int pte_present(pte_t pte)
 {
-	return (pte_val(pte) & _PAGE_INVALID) && !(pte_val(pte) & _PAGE_SWT);
+	/* Bit pattern: (pte & 0x001) == 0x001 */
+	return (pte_val(pte) & _PAGE_PRESENT) != 0;
 }
 
-static inline int pte_present(pte_t pte)
+static inline int pte_none(pte_t pte)
 {
-	unsigned long mask = _PAGE_RO | _PAGE_INVALID | _PAGE_SWT | _PAGE_SWX;
-	return (pte_val(pte) & mask) == _PAGE_TYPE_NONE ||
-		(!(pte_val(pte) & _PAGE_INVALID) &&
-		 !(pte_val(pte) & _PAGE_SWT));
+	/* Bit pattern: pte == 0x400 */
+	return pte_val(pte) == _PAGE_INVALID;
 }
 
 static inline int pte_file(pte_t pte)
 {
-	unsigned long mask = _PAGE_RO | _PAGE_INVALID | _PAGE_SWT;
-	return (pte_val(pte) & mask) == _PAGE_TYPE_FILE;
+	/* Bit pattern: (pte & 0x601) == 0x600 */
+	return (pte_val(pte) & (_PAGE_INVALID | _PAGE_PROTECT | _PAGE_PRESENT))
+		== (_PAGE_INVALID | _PAGE_PROTECT);
 }
 
 static inline int pte_special(pte_t pte)
@@ -695,7 +680,7 @@ static inline pgste_t pgste_update_young(pte_t *ptep, pgste_t pgste)
 	/* Transfer referenced bit to kvm user bits and pte */
 	if (young) {
 		pgste_val(pgste) |= PGSTE_UR_BIT;
-		pte_val(*ptep) |= _PAGE_SWR;
+		pte_val(*ptep) |= _PAGE_YOUNG;
 	}
 #endif
 	return pgste;
@@ -723,13 +708,13 @@ static inline void pgste_set_key(pte_t *ptep, pgste_t pgste, pte_t entry)
 
 static inline void pgste_set_pte(pte_t *ptep, pte_t entry)
 {
-	if (!MACHINE_HAS_ESOP && (pte_val(entry) & _PAGE_SWW)) {
+	if (!MACHINE_HAS_ESOP && (pte_val(entry) & _PAGE_WRITE)) {
 		/*
 		 * Without enhanced suppression-on-protection force
 		 * the dirty bit on for all writable ptes.
 		 */
-		pte_val(entry) |= _PAGE_SWC;
-		pte_val(entry) &= ~_PAGE_RO;
+		pte_val(entry) |= _PAGE_DIRTY;
+		pte_val(entry) &= ~_PAGE_PROTECT;
 	}
 	*ptep = entry;
 }
@@ -841,18 +826,18 @@ static inline void set_pte_at(struct mm_struct *mm, unsigned long addr,
  */
 static inline int pte_write(pte_t pte)
 {
-	return (pte_val(pte) & _PAGE_SWW) != 0;
+	return (pte_val(pte) & _PAGE_WRITE) != 0;
 }
 
 static inline int pte_dirty(pte_t pte)
 {
-	return (pte_val(pte) & _PAGE_SWC) != 0;
+	return (pte_val(pte) & _PAGE_DIRTY) != 0;
 }
 
 static inline int pte_young(pte_t pte)
 {
 #ifdef CONFIG_PGSTE
-	if (pte_val(pte) & _PAGE_SWR)
+	if (pte_val(pte) & _PAGE_YOUNG)
 		return 1;
 #endif
 	return 0;
@@ -880,12 +865,12 @@ static inline void pud_clear(pud_t *pud)
 
 static inline void pmd_clear(pmd_t *pmdp)
 {
-	pmd_val(*pmdp) = _SEGMENT_ENTRY_EMPTY;
+	pmd_val(*pmdp) = _SEGMENT_ENTRY_INVALID;
 }
 
 static inline void pte_clear(struct mm_struct *mm, unsigned long addr, pte_t *ptep)
 {
-	pte_val(*ptep) = _PAGE_TYPE_EMPTY;
+	pte_val(*ptep) = _PAGE_INVALID;
 }
 
 /*
@@ -896,49 +881,45 @@ static inline pte_t pte_modify(pte_t pte, pgprot_t newprot)
 {
 	pte_val(pte) &= _PAGE_CHG_MASK;
 	pte_val(pte) |= pgprot_val(newprot);
-	if ((pte_val(pte) & _PAGE_SWC) && (pte_val(pte) & _PAGE_SWW))
-		pte_val(pte) &= ~_PAGE_RO;
+	if ((pte_val(pte) & _PAGE_DIRTY) && (pte_val(pte) & _PAGE_WRITE))
+		pte_val(pte) &= ~_PAGE_PROTECT;
 	return pte;
 }
 
 static inline pte_t pte_wrprotect(pte_t pte)
 {
-	pte_val(pte) &= ~_PAGE_SWW;
-	/* Do not clobber _PAGE_TYPE_NONE pages!  */
-	if (!(pte_val(pte) & _PAGE_INVALID))
-		pte_val(pte) |= _PAGE_RO;
+	pte_val(pte) &= ~_PAGE_WRITE;
+	pte_val(pte) |= _PAGE_PROTECT;
 	return pte;
 }
 
 static inline pte_t pte_mkwrite(pte_t pte)
 {
-	pte_val(pte) |= _PAGE_SWW;
-	if (pte_val(pte) & _PAGE_SWC)
-		pte_val(pte) &= ~_PAGE_RO;
+	pte_val(pte) |= _PAGE_WRITE;
+	if (pte_val(pte) & _PAGE_DIRTY)
+		pte_val(pte) &= ~_PAGE_PROTECT;
 	return pte;
 }
 
 static inline pte_t pte_mkclean(pte_t pte)
 {
-	pte_val(pte) &= ~_PAGE_SWC;
-	/* Do not clobber _PAGE_TYPE_NONE pages!  */
-	if (!(pte_val(pte) & _PAGE_INVALID))
-		pte_val(pte) |= _PAGE_RO;
+	pte_val(pte) &= ~_PAGE_DIRTY;
+	pte_val(pte) |= _PAGE_PROTECT;
 	return pte;
 }
 
 static inline pte_t pte_mkdirty(pte_t pte)
 {
-	pte_val(pte) |= _PAGE_SWC;
-	if (pte_val(pte) & _PAGE_SWW)
-		pte_val(pte) &= ~_PAGE_RO;
+	pte_val(pte) |= _PAGE_DIRTY;
+	if (pte_val(pte) & _PAGE_WRITE)
+		pte_val(pte) &= ~_PAGE_PROTECT;
 	return pte;
 }
 
 static inline pte_t pte_mkold(pte_t pte)
 {
 #ifdef CONFIG_PGSTE
-	pte_val(pte) &= ~_PAGE_SWR;
+	pte_val(pte) &= ~_PAGE_YOUNG;
 #endif
 	return pte;
 }
@@ -957,7 +938,7 @@ static inline pte_t pte_mkspecial(pte_t pte)
 #ifdef CONFIG_HUGETLB_PAGE
 static inline pte_t pte_mkhuge(pte_t pte)
 {
-	pte_val(pte) |= (_SEGMENT_ENTRY_LARGE | _SEGMENT_ENTRY_CO);
+	pte_val(pte) |= _PAGE_LARGE;
 	return pte;
 }
 #endif
@@ -1076,7 +1057,7 @@ static inline pte_t ptep_get_and_clear(struct mm_struct *mm,
 	pte = *ptep;
 	if (!mm_exclusive(mm))
 		__ptep_ipte(address, ptep);
-	pte_val(*ptep) = _PAGE_TYPE_EMPTY;
+	pte_val(*ptep) = _PAGE_INVALID;
 
 	if (mm_has_pgste(mm)) {
 		pgste = pgste_update_all(&pte, pgste);
@@ -1139,7 +1120,7 @@ static inline pte_t ptep_clear_flush(struct vm_area_struct *vma,
 
 	pte = *ptep;
 	__ptep_ipte(address, ptep);
-	pte_val(*ptep) = _PAGE_TYPE_EMPTY;
+	pte_val(*ptep) = _PAGE_INVALID;
 
 	if (mm_has_pgste(vma->vm_mm)) {
 		pgste = pgste_update_all(&pte, pgste);
@@ -1172,7 +1153,7 @@ static inline pte_t ptep_get_and_clear_full(struct mm_struct *mm,
 	pte = *ptep;
 	if (!full)
 		__ptep_ipte(address, ptep);
-	pte_val(*ptep) = _PAGE_TYPE_EMPTY;
+	pte_val(*ptep) = _PAGE_INVALID;
 
 	if (mm_has_pgste(mm)) {
 		pgste = pgste_update_all(&pte, pgste);
@@ -1248,10 +1229,8 @@ static inline pte_t mk_pte(struct page *page, pgprot_t pgprot)
 	unsigned long physpage = page_to_phys(page);
 	pte_t __pte = mk_pte_phys(physpage, pgprot);
 
-	if ((pte_val(__pte) & _PAGE_SWW) && PageDirty(page)) {
-		pte_val(__pte) |= _PAGE_SWC;
-		pte_val(__pte) &= ~_PAGE_RO;
-	}
+	if (pte_write(__pte) && PageDirty(page))
+		__pte = pte_mkdirty(__pte);
 	return __pte;
 }
 
@@ -1313,7 +1292,7 @@ static inline void __pmd_idte(unsigned long address, pmd_t *pmdp)
 	unsigned long sto = (unsigned long) pmdp -
 			    pmd_index(address) * sizeof(pmd_t);
 
-	if (!(pmd_val(*pmdp) & _SEGMENT_ENTRY_INV)) {
+	if (!(pmd_val(*pmdp) & _SEGMENT_ENTRY_INVALID)) {
 		asm volatile(
 			"	.insn	rrf,0xb98e0000,%2,%3,0,0"
 			: "=m" (*pmdp)
@@ -1324,18 +1303,31 @@ static inline void __pmd_idte(unsigned long address, pmd_t *pmdp)
 	}
 }
 
+static inline void __pmd_csp(pmd_t *pmdp)
+{
+	register unsigned long reg2 asm("2") = pmd_val(*pmdp);
+	register unsigned long reg3 asm("3") = pmd_val(*pmdp) |
+					       _SEGMENT_ENTRY_INVALID;
+	register unsigned long reg4 asm("4") = ((unsigned long) pmdp) + 5;
+
+	asm volatile(
+		"	csp %1,%3"
+		: "=m" (*pmdp)
+		: "d" (reg2), "d" (reg3), "d" (reg4), "m" (*pmdp) : "cc");
+}
+
 #if defined(CONFIG_TRANSPARENT_HUGEPAGE) || defined(CONFIG_HUGETLB_PAGE)
 static inline unsigned long massage_pgprot_pmd(pgprot_t pgprot)
 {
 	/*
-	 * pgprot is PAGE_NONE, PAGE_RO, or PAGE_RW (see __Pxxx / __Sxxx)
+	 * pgprot is PAGE_NONE, PAGE_READ, or PAGE_WRITE (see __Pxxx / __Sxxx)
 	 * Convert to segment table entry format.
 	 */
 	if (pgprot_val(pgprot) == pgprot_val(PAGE_NONE))
 		return pgprot_val(SEGMENT_NONE);
-	if (pgprot_val(pgprot) == pgprot_val(PAGE_RO))
-		return pgprot_val(SEGMENT_RO);
-	return pgprot_val(SEGMENT_RW);
+	if (pgprot_val(pgprot) == pgprot_val(PAGE_READ))
+		return pgprot_val(SEGMENT_READ);
+	return pgprot_val(SEGMENT_WRITE);
 }
 
 static inline pmd_t pmd_modify(pmd_t pmd, pgprot_t newprot)
@@ -1354,9 +1346,9 @@ static inline pmd_t mk_pmd_phys(unsigned long physpage, pgprot_t pgprot)
 
 static inline pmd_t pmd_mkwrite(pmd_t pmd)
 {
-	/* Do not clobber _HPAGE_TYPE_NONE pages! */
-	if (!(pmd_val(pmd) & _SEGMENT_ENTRY_INV))
-		pmd_val(pmd) &= ~_SEGMENT_ENTRY_RO;
+	/* Do not clobber PROT_NONE pages! */
+	if (!(pmd_val(pmd) & _SEGMENT_ENTRY_INVALID))
+		pmd_val(pmd) &= ~_SEGMENT_ENTRY_PROTECT;
 	return pmd;
 }
 #endif /* CONFIG_TRANSPARENT_HUGEPAGE || CONFIG_HUGETLB_PAGE */
@@ -1378,7 +1370,7 @@ static inline int pmd_trans_splitting(pmd_t pmd)
 static inline void set_pmd_at(struct mm_struct *mm, unsigned long addr,
 			      pmd_t *pmdp, pmd_t entry)
 {
-	if (!(pmd_val(entry) & _SEGMENT_ENTRY_INV) && MACHINE_HAS_EDAT1)
+	if (!(pmd_val(entry) & _SEGMENT_ENTRY_INVALID) && MACHINE_HAS_EDAT1)
 		pmd_val(entry) |= _SEGMENT_ENTRY_CO;
 	*pmdp = entry;
 }
@@ -1391,7 +1383,7 @@ static inline pmd_t pmd_mkhuge(pmd_t pmd)
 
 static inline pmd_t pmd_wrprotect(pmd_t pmd)
 {
-	pmd_val(pmd) |= _SEGMENT_ENTRY_RO;
+	pmd_val(pmd) |= _SEGMENT_ENTRY_PROTECT;
 	return pmd;
 }
 
@@ -1510,10 +1502,8 @@ static inline unsigned long pmd_pfn(pmd_t pmd)
  * exception will occur instead of a page translation exception. The
  * specifiation exception has the bad habit not to store necessary
  * information in the lowcore.
- * Bit 21 and bit 22 are the page invalid bit and the page protection
- * bit. We set both to indicate a swapped page.
- * Bit 30 and 31 are used to distinguish the different page types. For
- * a swapped page these bits need to be zero.
+ * Bits 21, 22, 30 and 31 are used to indicate the page type.
+ * A swap pte is indicated by bit pattern (pte & 0x603) == 0x402
  * This leaves the bits 1-19 and bits 24-29 to store type and offset.
  * We use the 5 bits from 25-29 for the type and the 20 bits from 1-19
  * plus 24 for the offset.
@@ -1527,10 +1517,8 @@ static inline unsigned long pmd_pfn(pmd_t pmd)
  * exception will occur instead of a page translation exception. The
  * specifiation exception has the bad habit not to store necessary
  * information in the lowcore.
- * Bit 53 and bit 54 are the page invalid bit and the page protection
- * bit. We set both to indicate a swapped page.
- * Bit 62 and 63 are used to distinguish the different page types. For
- * a swapped page these bits need to be zero.
+ * Bits 53, 54, 62 and 63 are used to indicate the page type.
+ * A swap pte is indicated by bit pattern (pte & 0x603) == 0x402
  * This leaves the bits 0-51 and bits 56-61 to store type and offset.
  * We use the 5 bits from 57-61 for the type and the 53 bits from 0-51
  * plus 56 for the offset.
@@ -1547,7 +1535,7 @@ static inline pte_t mk_swap_pte(unsigned long type, unsigned long offset)
 {
 	pte_t pte;
 	offset &= __SWP_OFFSET_MASK;
-	pte_val(pte) = _PAGE_TYPE_SWAP | ((type & 0x1f) << 2) |
+	pte_val(pte) = _PAGE_INVALID | _PAGE_TYPE | ((type & 0x1f) << 2) |
 		((offset & 1UL) << 7) | ((offset & ~1UL) << 11);
 	return pte;
 }
@@ -1570,7 +1558,7 @@ static inline pte_t mk_swap_pte(unsigned long type, unsigned long offset)
 
 #define pgoff_to_pte(__off) \
 	((pte_t) { ((((__off) & 0x7f) << 1) + (((__off) >> 7) << 12)) \
-		   | _PAGE_TYPE_FILE })
+		   | _PAGE_INVALID | _PAGE_PROTECT })
 
 #endif /* !__ASSEMBLY__ */
 

arch/s390/kernel/vdso.c

@@ -113,11 +113,11 @@ int vdso_alloc_per_cpu(struct _lowcore *lowcore)
 
 	clear_table((unsigned long *) segment_table, _SEGMENT_ENTRY_EMPTY,
 		    PAGE_SIZE << SEGMENT_ORDER);
-	clear_table((unsigned long *) page_table, _PAGE_TYPE_EMPTY,
+	clear_table((unsigned long *) page_table, _PAGE_INVALID,
 		    256*sizeof(unsigned long));
 
 	*(unsigned long *) segment_table = _SEGMENT_ENTRY + page_table;
-	*(unsigned long *) page_table = _PAGE_RO + page_frame;
+	*(unsigned long *) page_table = _PAGE_PROTECT + page_frame;
 
 	psal = (u32 *) (page_table + 256*sizeof(unsigned long));
 	aste = psal + 32;

arch/s390/lib/uaccess_pt.c

@@ -86,28 +86,28 @@ static unsigned long follow_table(struct mm_struct *mm,
 	switch (mm->context.asce_bits & _ASCE_TYPE_MASK) {
 	case _ASCE_TYPE_REGION1:
 		table = table + ((address >> 53) & 0x7ff);
-		if (unlikely(*table & _REGION_ENTRY_INV))
+		if (unlikely(*table & _REGION_ENTRY_INVALID))
 			return -0x39UL;
 		table = (unsigned long *)(*table & _REGION_ENTRY_ORIGIN);
 		/* fallthrough */
 	case _ASCE_TYPE_REGION2:
 		table = table + ((address >> 42) & 0x7ff);
-		if (unlikely(*table & _REGION_ENTRY_INV))
+		if (unlikely(*table & _REGION_ENTRY_INVALID))
 			return -0x3aUL;
 		table = (unsigned long *)(*table & _REGION_ENTRY_ORIGIN);
 		/* fallthrough */
 	case _ASCE_TYPE_REGION3:
 		table = table + ((address >> 31) & 0x7ff);
-		if (unlikely(*table & _REGION_ENTRY_INV))
+		if (unlikely(*table & _REGION_ENTRY_INVALID))
 			return -0x3bUL;
 		table = (unsigned long *)(*table & _REGION_ENTRY_ORIGIN);
 		/* fallthrough */
 	case _ASCE_TYPE_SEGMENT:
 		table = table + ((address >> 20) & 0x7ff);
-		if (unlikely(*table & _SEGMENT_ENTRY_INV))
+		if (unlikely(*table & _SEGMENT_ENTRY_INVALID))
 			return -0x10UL;
 		if (unlikely(*table & _SEGMENT_ENTRY_LARGE)) {
-			if (write && (*table & _SEGMENT_ENTRY_RO))
+			if (write && (*table & _SEGMENT_ENTRY_PROTECT))
 				return -0x04UL;
 			return (*table & _SEGMENT_ENTRY_ORIGIN_LARGE) +
 				(address & ~_SEGMENT_ENTRY_ORIGIN_LARGE);
@@ -117,7 +117,7 @@ static unsigned long follow_table(struct mm_struct *mm,
 	table = table + ((address >> 12) & 0xff);
 	if (unlikely(*table & _PAGE_INVALID))
 		return -0x11UL;
-	if (write && (*table & _PAGE_RO))
+	if (write && (*table & _PAGE_PROTECT))
 		return -0x04UL;
 	return (*table & PAGE_MASK) + (address & ~PAGE_MASK);
 }
@@ -130,13 +130,13 @@ static unsigned long follow_table(struct mm_struct *mm,
 	unsigned long *table = (unsigned long *)__pa(mm->pgd);
 
 	table = table + ((address >> 20) & 0x7ff);
-	if (unlikely(*table & _SEGMENT_ENTRY_INV))
+	if (unlikely(*table & _SEGMENT_ENTRY_INVALID))
 		return -0x10UL;
 	table = (unsigned long *)(*table & _SEGMENT_ENTRY_ORIGIN);
 	table = table + ((address >> 12) & 0xff);
 	if (unlikely(*table & _PAGE_INVALID))
 		return -0x11UL;
-	if (write && (*table & _PAGE_RO))
+	if (write && (*table & _PAGE_PROTECT))
 		return -0x04UL;
 	return (*table & PAGE_MASK) + (address & ~PAGE_MASK);
 }

arch/s390/mm/dump_pagetables.c

@@ -53,7 +53,7 @@ static void print_prot(struct seq_file *m, unsigned int pr, int level)
 		seq_printf(m, "I\n");
 		return;
 	}
-	seq_printf(m, "%s", pr & _PAGE_RO ? "RO " : "RW ");
+	seq_printf(m, "%s", pr & _PAGE_PROTECT ? "RO " : "RW ");
 	seq_printf(m, "%s", pr & _PAGE_CO ? "CO " : "   ");
 	seq_putc(m, '\n');
 }
@@ -105,12 +105,12 @@ static void note_page(struct seq_file *m, struct pg_state *st,
 }
 
 /*
- * The actual page table walker functions. In order to keep the implementation
- * of print_prot() short, we only check and pass _PAGE_INVALID and _PAGE_RO
- * flags to note_page() if a region, segment or page table entry is invalid or
- * read-only.
- * After all it's just a hint that the current level being walked contains an
- * invalid or read-only entry.
+ * The actual page table walker functions. In order to keep the
+ * implementation of print_prot() short, we only check and pass
+ * _PAGE_INVALID and _PAGE_PROTECT flags to note_page() if a region,
+ * segment or page table entry is invalid or read-only.
+ * After all it's just a hint that the current level being walked
+ * contains an invalid or read-only entry.
  */
 static void walk_pte_level(struct seq_file *m, struct pg_state *st,
 			   pmd_t *pmd, unsigned long addr)
@@ -122,14 +122,14 @@ static void walk_pte_level(struct seq_file *m, struct pg_state *st,
 	for (i = 0; i < PTRS_PER_PTE && addr < max_addr; i++) {
 		st->current_address = addr;
 		pte = pte_offset_kernel(pmd, addr);
-		prot = pte_val(*pte) & (_PAGE_RO | _PAGE_INVALID);
+		prot = pte_val(*pte) & (_PAGE_PROTECT | _PAGE_INVALID);
 		note_page(m, st, prot, 4);
 		addr += PAGE_SIZE;
 	}
 }
 
 #ifdef CONFIG_64BIT
-#define _PMD_PROT_MASK (_SEGMENT_ENTRY_RO | _SEGMENT_ENTRY_CO)
+#define _PMD_PROT_MASK (_SEGMENT_ENTRY_PROTECT | _SEGMENT_ENTRY_CO)
 #else
 #define _PMD_PROT_MASK 0
 #endif

arch/s390/mm/gup.c

@@ -24,7 +24,7 @@ static inline int gup_pte_range(pmd_t *pmdp, pmd_t pmd, unsigned long addr,
 	pte_t *ptep, pte;
 	struct page *page;
 
-	mask = (write ? _PAGE_RO : 0) | _PAGE_INVALID | _PAGE_SPECIAL;
+	mask = (write ? _PAGE_PROTECT : 0) | _PAGE_INVALID | _PAGE_SPECIAL;
 
 	ptep = ((pte_t *) pmd_deref(pmd)) + pte_index(addr);
 	do {
@@ -55,8 +55,8 @@ static inline int gup_huge_pmd(pmd_t *pmdp, pmd_t pmd, unsigned long addr,
 	struct page *head, *page, *tail;
 	int refs;
 
-	result = write ? 0 : _SEGMENT_ENTRY_RO;
-	mask = result | _SEGMENT_ENTRY_INV;
+	result = write ? 0 : _SEGMENT_ENTRY_PROTECT;
+	mask = result | _SEGMENT_ENTRY_INVALID;
 	if ((pmd_val(pmd) & mask) != result)
 		return 0;
 	VM_BUG_ON(!pfn_valid(pmd_val(pmd) >> PAGE_SHIFT));

arch/s390/mm/hugetlbpage.c

@@ -8,21 +8,107 @@
 #include <linux/mm.h>
 #include <linux/hugetlb.h>
 
+static inline pmd_t __pte_to_pmd(pte_t pte)
+{
+	int none, prot;
+	pmd_t pmd;
+
+	/*
+	 * Convert encoding	  pte bits	  pmd bits
+	 *			.IR.....wdtp	..R...I.....
+	 * empty		.10.....0000 -> ..0...1.....
+	 * prot-none, clean	.11.....0001 -> ..1...1.....
+	 * prot-none, dirty	.10.....0101 -> ..1...1.....
+	 * read-only, clean	.01.....0001 -> ..1...0.....
+	 * read-only, dirty	.01.....0101 -> ..1...0.....
+	 * read-write, clean	.01.....1001 -> ..0...0.....
+	 * read-write, dirty	.00.....1101 -> ..0...0.....
+	 * Huge ptes are dirty by definition, a clean pte is made dirty
+	 * by the conversion.
+	 */
+	if (pte_present(pte)) {
+		pmd_val(pmd) = pte_val(pte) & PAGE_MASK;
+		if (pte_val(pte) & _PAGE_INVALID)
+			pmd_val(pmd) |= _SEGMENT_ENTRY_INVALID;
+		none = (pte_val(pte) & _PAGE_PRESENT) &&
+			(pte_val(pte) & _PAGE_INVALID);
+		prot = (pte_val(pte) & _PAGE_PROTECT);
+		if (prot || none)
+			pmd_val(pmd) |= _SEGMENT_ENTRY_PROTECT;
+	} else
+		pmd_val(pmd) = _SEGMENT_ENTRY_INVALID;
+	return pmd;
+}
+
+static inline pte_t __pmd_to_pte(pmd_t pmd)
+{
+	pte_t pte;
+
+	/*
+	 * Convert encoding	  pmd bits	  pte bits
+	 *			..R...I.....	.IR.....wdtp
+	 * empty		..0...1..... -> .10.....0000
+	 * prot-none, young	..1...1..... -> .10.....0101
+	 * read-only, young	..1...0..... -> .01.....0101
+	 * read-write, young	..0...0..... -> .00.....1101
+	 * Huge ptes are dirty by definition
+	 */
+	if (pmd_present(pmd)) {
+		pte_val(pte) = _PAGE_PRESENT | _PAGE_LARGE | _PAGE_DIRTY |
+			(pmd_val(pmd) & PAGE_MASK);
+		if (pmd_val(pmd) & _SEGMENT_ENTRY_INVALID)
+			pte_val(pte) |= _PAGE_INVALID;
+		else {
+			if (pmd_val(pmd) & _SEGMENT_ENTRY_PROTECT)
+				pte_val(pte) |= _PAGE_PROTECT;
+			else
+				pte_val(pte) |= _PAGE_WRITE;
+		}
+	} else
+		pte_val(pte) = _PAGE_INVALID;
+	return pte;
+}
 
 void set_huge_pte_at(struct mm_struct *mm, unsigned long addr,
-				   pte_t *pteptr, pte_t pteval)
+		     pte_t *ptep, pte_t pte)
 {
-	pmd_t *pmdp = (pmd_t *) pteptr;
-	unsigned long mask;
+	pmd_t pmd;
 
+	pmd = __pte_to_pmd(pte);
 	if (!MACHINE_HAS_HPAGE) {
-		pteptr = (pte_t *) pte_page(pteval)[1].index;
-		mask = pte_val(pteval) &
-				(_SEGMENT_ENTRY_INV | _SEGMENT_ENTRY_RO);
-		pte_val(pteval) = (_SEGMENT_ENTRY + __pa(pteptr)) | mask;
+		pmd_val(pmd) &= ~_SEGMENT_ENTRY_ORIGIN;
+		pmd_val(pmd) |= pte_page(pte)[1].index;
+	} else
+		pmd_val(pmd) |= _SEGMENT_ENTRY_LARGE | _SEGMENT_ENTRY_CO;
+	*(pmd_t *) ptep = pmd;
+}
+
+pte_t huge_ptep_get(pte_t *ptep)
+{
+	unsigned long origin;
+	pmd_t pmd;
+
+	pmd = *(pmd_t *) ptep;
+	if (!MACHINE_HAS_HPAGE && pmd_present(pmd)) {
+		origin = pmd_val(pmd) & _SEGMENT_ENTRY_ORIGIN;
+		pmd_val(pmd) &= ~_SEGMENT_ENTRY_ORIGIN;
+		pmd_val(pmd) |= *(unsigned long *) origin;
 	}
+	return __pmd_to_pte(pmd);
+}
 
-	pmd_val(*pmdp) = pte_val(pteval);
+pte_t huge_ptep_get_and_clear(struct mm_struct *mm,
+			      unsigned long addr, pte_t *ptep)
+{
+	pmd_t *pmdp = (pmd_t *) ptep;
+	pte_t pte = huge_ptep_get(ptep);
+
+	if (MACHINE_HAS_IDTE)
+		__pmd_idte(addr, pmdp);
+	else
+		__pmd_csp(pmdp);
+	pmd_val(*pmdp) = _SEGMENT_ENTRY_EMPTY;
+	return pte;
 }
 
 int arch_prepare_hugepage(struct page *page)
@@ -58,7 +144,7 @@ void arch_release_hugepage(struct page *page)
 	ptep = (pte_t *) page[1].index;
 	if (!ptep)
 		return;
-	clear_table((unsigned long *) ptep, _PAGE_TYPE_EMPTY,
+	clear_table((unsigned long *) ptep, _PAGE_INVALID,
 		    PTRS_PER_PTE * sizeof(pte_t));
 	page_table_free(&init_mm, (unsigned long *) ptep);
 	page[1].index = 0;

arch/s390/mm/pageattr.c

@@ -118,7 +118,7 @@ void kernel_map_pages(struct page *page, int numpages, int enable)
 		pte = pte_offset_kernel(pmd, address);
 		if (!enable) {
 			__ptep_ipte(address, pte);
-			pte_val(*pte) = _PAGE_TYPE_EMPTY;
+			pte_val(*pte) = _PAGE_INVALID;
 			continue;
 		}
 		pte_val(*pte) = __pa(address);

arch/s390/mm/pgtable.c

@@ -161,7 +161,7 @@ static int gmap_unlink_segment(struct gmap *gmap, unsigned long *table)
 	struct gmap_rmap *rmap;
 	struct page *page;
 
-	if (*table & _SEGMENT_ENTRY_INV)
+	if (*table & _SEGMENT_ENTRY_INVALID)
 		return 0;
 	page = pfn_to_page(*table >> PAGE_SHIFT);
 	mp = (struct gmap_pgtable *) page->index;
@@ -172,7 +172,7 @@ static int gmap_unlink_segment(struct gmap *gmap, unsigned long *table)
 		kfree(rmap);
 		break;
 	}
-	*table = _SEGMENT_ENTRY_INV | _SEGMENT_ENTRY_RO | mp->vmaddr;
+	*table = mp->vmaddr | _SEGMENT_ENTRY_INVALID | _SEGMENT_ENTRY_PROTECT;
 	return 1;
 }
 
@@ -258,7 +258,7 @@ static int gmap_alloc_table(struct gmap *gmap,
 		return -ENOMEM;
 	new = (unsigned long *) page_to_phys(page);
 	crst_table_init(new, init);
-	if (*table & _REGION_ENTRY_INV) {
+	if (*table & _REGION_ENTRY_INVALID) {
 		list_add(&page->lru, &gmap->crst_list);
 		*table = (unsigned long) new | _REGION_ENTRY_LENGTH |
 			(*table & _REGION_ENTRY_TYPE_MASK);
@@ -292,22 +292,22 @@ int gmap_unmap_segment(struct gmap *gmap, unsigned long to, unsigned long len)
 	for (off = 0; off < len; off += PMD_SIZE) {
 		/* Walk the guest addr space page table */
 		table = gmap->table + (((to + off) >> 53) & 0x7ff);
-		if (*table & _REGION_ENTRY_INV)
+		if (*table & _REGION_ENTRY_INVALID)
 			goto out;
 		table = (unsigned long *)(*table & _REGION_ENTRY_ORIGIN);
 		table = table + (((to + off) >> 42) & 0x7ff);
-		if (*table & _REGION_ENTRY_INV)
+		if (*table & _REGION_ENTRY_INVALID)
 			goto out;
 		table = (unsigned long *)(*table & _REGION_ENTRY_ORIGIN);
 		table = table + (((to + off) >> 31) & 0x7ff);
-		if (*table & _REGION_ENTRY_INV)
+		if (*table & _REGION_ENTRY_INVALID)
 			goto out;
 		table = (unsigned long *)(*table & _REGION_ENTRY_ORIGIN);
 		table = table + (((to + off) >> 20) & 0x7ff);
 
 		/* Clear segment table entry in guest address space. */
 		flush |= gmap_unlink_segment(gmap, table);
-		*table = _SEGMENT_ENTRY_INV;
+		*table = _SEGMENT_ENTRY_INVALID;
 	}
 out:
 	spin_unlock(&gmap->mm->page_table_lock);
@@ -345,17 +345,17 @@ int gmap_map_segment(struct gmap *gmap, unsigned long from,
 	for (off = 0; off < len; off += PMD_SIZE) {
 		/* Walk the gmap address space page table */
 		table = gmap->table + (((to + off) >> 53) & 0x7ff);
-		if ((*table & _REGION_ENTRY_INV) &&
+		if ((*table & _REGION_ENTRY_INVALID) &&
 		    gmap_alloc_table(gmap, table, _REGION2_ENTRY_EMPTY))
 			goto out_unmap;
 		table = (unsigned long *)(*table & _REGION_ENTRY_ORIGIN);
 		table = table + (((to + off) >> 42) & 0x7ff);
-		if ((*table & _REGION_ENTRY_INV) &&
+		if ((*table & _REGION_ENTRY_INVALID) &&
 		    gmap_alloc_table(gmap, table, _REGION3_ENTRY_EMPTY))
 			goto out_unmap;
 		table = (unsigned long *)(*table & _REGION_ENTRY_ORIGIN);
 		table = table + (((to + off) >> 31) & 0x7ff);
-		if ((*table & _REGION_ENTRY_INV) &&
+		if ((*table & _REGION_ENTRY_INVALID) &&
 		    gmap_alloc_table(gmap, table, _SEGMENT_ENTRY_EMPTY))
 			goto out_unmap;
 		table = (unsigned long *) (*table & _REGION_ENTRY_ORIGIN);
@@ -363,7 +363,8 @@ int gmap_map_segment(struct gmap *gmap, unsigned long from,
 
 		/* Store 'from' address in an invalid segment table entry. */
 		flush |= gmap_unlink_segment(gmap, table);
-		*table = _SEGMENT_ENTRY_INV | _SEGMENT_ENTRY_RO | (from + off);
+		*table =  (from + off) | (_SEGMENT_ENTRY_INVALID |
+					  _SEGMENT_ENTRY_PROTECT);
 	}
 	spin_unlock(&gmap->mm->page_table_lock);
 	up_read(&gmap->mm->mmap_sem);
@@ -384,15 +385,15 @@ static unsigned long *gmap_table_walk(unsigned long address, struct gmap *gmap)
 	unsigned long *table;
 
 	table = gmap->table + ((address >> 53) & 0x7ff);
-	if (unlikely(*table & _REGION_ENTRY_INV))
+	if (unlikely(*table & _REGION_ENTRY_INVALID))
 		return ERR_PTR(-EFAULT);
 	table = (unsigned long *)(*table & _REGION_ENTRY_ORIGIN);
 	table = table + ((address >> 42) & 0x7ff);
-	if (unlikely(*table & _REGION_ENTRY_INV))
+	if (unlikely(*table & _REGION_ENTRY_INVALID))
 		return ERR_PTR(-EFAULT);
 	table = (unsigned long *)(*table & _REGION_ENTRY_ORIGIN);
 	table = table + ((address >> 31) & 0x7ff);
-	if (unlikely(*table & _REGION_ENTRY_INV))
+	if (unlikely(*table & _REGION_ENTRY_INVALID))
 		return ERR_PTR(-EFAULT);
 	table = (unsigned long *)(*table & _REGION_ENTRY_ORIGIN);
 	table = table + ((address >> 20) & 0x7ff);
@@ -422,11 +423,11 @@ unsigned long __gmap_translate(unsigned long address, struct gmap *gmap)
 		return PTR_ERR(segment_ptr);
 	/* Convert the gmap address to an mm address. */
 	segment = *segment_ptr;
-	if (!(segment & _SEGMENT_ENTRY_INV)) {
+	if (!(segment & _SEGMENT_ENTRY_INVALID)) {
 		page = pfn_to_page(segment >> PAGE_SHIFT);
 		mp = (struct gmap_pgtable *) page->index;
 		return mp->vmaddr | (address & ~PMD_MASK);
-	} else if (segment & _SEGMENT_ENTRY_RO) {
+	} else if (segment & _SEGMENT_ENTRY_PROTECT) {
 		vmaddr = segment & _SEGMENT_ENTRY_ORIGIN;
 		return vmaddr | (address & ~PMD_MASK);
 	}
@@ -517,8 +518,8 @@ static void gmap_disconnect_pgtable(struct mm_struct *mm, unsigned long *table)
 	page = pfn_to_page(__pa(table) >> PAGE_SHIFT);
 	mp = (struct gmap_pgtable *) page->index;
 	list_for_each_entry_safe(rmap, next, &mp->mapper, list) {
-		*rmap->entry =
-			_SEGMENT_ENTRY_INV | _SEGMENT_ENTRY_RO | mp->vmaddr;
+		*rmap->entry = mp->vmaddr | (_SEGMENT_ENTRY_INVALID |
+					     _SEGMENT_ENTRY_PROTECT);
 		list_del(&rmap->list);
 		kfree(rmap);
 		flush = 1;
@@ -545,13 +546,13 @@ unsigned long __gmap_fault(unsigned long address, struct gmap *gmap)
 	/* Convert the gmap address to an mm address. */
 	while (1) {
 		segment = *segment_ptr;
-		if (!(segment & _SEGMENT_ENTRY_INV)) {
+		if (!(segment & _SEGMENT_ENTRY_INVALID)) {
 			/* Page table is present */
 			page = pfn_to_page(segment >> PAGE_SHIFT);
 			mp = (struct gmap_pgtable *) page->index;
 			return mp->vmaddr | (address & ~PMD_MASK);
 		}
-		if (!(segment & _SEGMENT_ENTRY_RO))
+		if (!(segment & _SEGMENT_ENTRY_PROTECT))
 			/* Nothing mapped in the gmap address space. */
 			break;
 		rc = gmap_connect_pgtable(address, segment, segment_ptr, gmap);
@@ -586,25 +587,25 @@ void gmap_discard(unsigned long from, unsigned long to, struct gmap *gmap)
 	while (address < to) {
 		/* Walk the gmap address space page table */
 		table = gmap->table + ((address >> 53) & 0x7ff);
-		if (unlikely(*table & _REGION_ENTRY_INV)) {
+		if (unlikely(*table & _REGION_ENTRY_INVALID)) {
 			address = (address + PMD_SIZE) & PMD_MASK;
 			continue;
 		}
 		table = (unsigned long *)(*table & _REGION_ENTRY_ORIGIN);
 		table = table + ((address >> 42) & 0x7ff);
-		if (unlikely(*table & _REGION_ENTRY_INV)) {
+		if (unlikely(*table & _REGION_ENTRY_INVALID)) {
 			address = (address + PMD_SIZE) & PMD_MASK;
 			continue;
 		}
 		table = (unsigned long *)(*table & _REGION_ENTRY_ORIGIN);
 		table = table + ((address >> 31) & 0x7ff);
-		if (unlikely(*table & _REGION_ENTRY_INV)) {
+		if (unlikely(*table & _REGION_ENTRY_INVALID)) {
 			address = (address + PMD_SIZE) & PMD_MASK;
 			continue;
 		}
 		table = (unsigned long *)(*table & _REGION_ENTRY_ORIGIN);
 		table = table + ((address >> 20) & 0x7ff);
-		if (unlikely(*table & _SEGMENT_ENTRY_INV)) {
+		if (unlikely(*table & _SEGMENT_ENTRY_INVALID)) {
 			address = (address + PMD_SIZE) & PMD_MASK;
 			continue;
 		}
@@ -687,7 +688,7 @@ int gmap_ipte_notify(struct gmap *gmap, unsigned long start, unsigned long len)
 			continue;
 		/* Set notification bit in the pgste of the pte */
 		entry = *ptep;
-		if ((pte_val(entry) & (_PAGE_INVALID | _PAGE_RO)) == 0) {
+		if ((pte_val(entry) & (_PAGE_INVALID | _PAGE_PROTECT)) == 0) {
 			pgste = pgste_get_lock(ptep);
 			pgste_val(pgste) |= PGSTE_IN_BIT;
 			pgste_set_unlock(ptep, pgste);
@@ -752,7 +753,7 @@ static inline unsigned long *page_table_alloc_pgste(struct mm_struct *mm,
 	page->index = (unsigned long) mp;
 	atomic_set(&page->_mapcount, 3);
 	table = (unsigned long *) page_to_phys(page);
-	clear_table(table, _PAGE_TYPE_EMPTY, PAGE_SIZE/2);
+	clear_table(table, _PAGE_INVALID, PAGE_SIZE/2);
 	clear_table(table + PTRS_PER_PTE, 0, PAGE_SIZE/2);
 	return table;
 }
@@ -878,7 +879,7 @@ unsigned long *page_table_alloc(struct mm_struct *mm, unsigned long vmaddr)
 		pgtable_page_ctor(page);
 		atomic_set(&page->_mapcount, 1);
 		table = (unsigned long *) page_to_phys(page);
-		clear_table(table, _PAGE_TYPE_EMPTY, PAGE_SIZE);
+		clear_table(table, _PAGE_INVALID, PAGE_SIZE);
 		spin_lock_bh(&mm->context.list_lock);
 		list_add(&page->lru, &mm->context.pgtable_list);
 	} else {
@@ -1198,9 +1199,9 @@ pgtable_t pgtable_trans_huge_withdraw(struct mm_struct *mm, pmd_t *pmdp)
 		list_del(lh);
 	}
 	ptep = (pte_t *) pgtable;
-	pte_val(*ptep) = _PAGE_TYPE_EMPTY;
+	pte_val(*ptep) = _PAGE_INVALID;
 	ptep++;
-	pte_val(*ptep) = _PAGE_TYPE_EMPTY;
+	pte_val(*ptep) = _PAGE_INVALID;
 	return pgtable;
 }
 #endif /* CONFIG_TRANSPARENT_HUGEPAGE */

arch/s390/mm/vmem.c

@@ -69,7 +69,7 @@ static pte_t __ref *vmem_pte_alloc(unsigned long address)
 		pte = alloc_bootmem(PTRS_PER_PTE * sizeof(pte_t));
 	if (!pte)
 		return NULL;
-	clear_table((unsigned long *) pte, _PAGE_TYPE_EMPTY,
+	clear_table((unsigned long *) pte, _PAGE_INVALID,
 		    PTRS_PER_PTE * sizeof(pte_t));
 	return pte;
 }
@@ -101,7 +101,7 @@ static int vmem_add_mem(unsigned long start, unsigned long size, int ro)
 		    !(address & ~PUD_MASK) && (address + PUD_SIZE <= end)) {
 			pud_val(*pu_dir) = __pa(address) |
 				_REGION_ENTRY_TYPE_R3 | _REGION3_ENTRY_LARGE |
-				(ro ? _REGION_ENTRY_RO : 0);
+				(ro ? _REGION_ENTRY_PROTECT : 0);
 			address += PUD_SIZE;
 			continue;
 		}
@@ -118,7 +118,7 @@ static int vmem_add_mem(unsigned long start, unsigned long size, int ro)
 		    !(address & ~PMD_MASK) && (address + PMD_SIZE <= end)) {
 			pmd_val(*pm_dir) = __pa(address) |
 				_SEGMENT_ENTRY | _SEGMENT_ENTRY_LARGE |
-				(ro ? _SEGMENT_ENTRY_RO : 0);
+				(ro ? _SEGMENT_ENTRY_PROTECT : 0);
 			address += PMD_SIZE;
 			continue;
 		}
@@ -131,7 +131,8 @@ static int vmem_add_mem(unsigned long start, unsigned long size, int ro)
 		}
 
 		pt_dir = pte_offset_kernel(pm_dir, address);
-		pte_val(*pt_dir) = __pa(address) | (ro ? _PAGE_RO : 0);
+		pte_val(*pt_dir) = __pa(address) |
+			pgprot_val(ro ? PAGE_KERNEL_RO : PAGE_KERNEL);
 		address += PAGE_SIZE;
 	}
 	ret = 0;
@@ -154,7 +155,7 @@ static void vmem_remove_range(unsigned long start, unsigned long size)
 	pte_t *pt_dir;
 	pte_t  pte;
 
-	pte_val(pte) = _PAGE_TYPE_EMPTY;
+	pte_val(pte) = _PAGE_INVALID;
 	while (address < end) {
 		pg_dir = pgd_offset_k(address);
 		if (pgd_none(*pg_dir)) {
@@ -255,7 +256,8 @@ int __meminit vmemmap_populate(unsigned long start, unsigned long end, int node)
 			new_page =__pa(vmem_alloc_pages(0));
 			if (!new_page)
 				goto out;
-			pte_val(*pt_dir) = __pa(new_page);
+			pte_val(*pt_dir) =
+				__pa(new_page) | pgprot_val(PAGE_KERNEL);
 		}
 		address += PAGE_SIZE;
 	}