Merge branch 'pgt' (early part) into devel

10 files changed, 128 insertions, 100 deletions

diff --git a/arch/arm/mm/Makefile b/arch/arm/mm/Makefile
index d63b6c413758..00d74a04af3a 100644
--- a/arch/arm/mm/Makefile
+++ b/arch/arm/mm/Makefile
@@ -5,8 +5,8 @@
 obj-y                           := dma-mapping.o extable.o fault.o init.o \
                                    iomap.o
 
-obj-$(CONFIG_MMU)               += fault-armv.o flush.o ioremap.o mmap.o \
-                                   pgd.o mmu.o vmregion.o
+obj-$(CONFIG_MMU)               += fault-armv.o flush.o idmap.o ioremap.o \
+                                   mmap.o pgd.o mmu.o vmregion.o
 
 ifneq ($(CONFIG_MMU),y)
 obj-y                           += nommu.o
diff --git a/arch/arm/mm/fault-armv.c b/arch/arm/mm/fault-armv.c
index 83e59f870426..01210dba0221 100644
--- a/arch/arm/mm/fault-armv.c
+++ b/arch/arm/mm/fault-armv.c
@@ -26,7 +26,7 @@
 
 #include "mm.h"
 
-static unsigned long shared_pte_mask = L_PTE_MT_BUFFERABLE;
+static pteval_t shared_pte_mask = L_PTE_MT_BUFFERABLE;
 
 #if __LINUX_ARM_ARCH__ < 6
 /*
diff --git a/arch/arm/mm/fault.c b/arch/arm/mm/fault.c
index 1e21e125fe3a..f10f9bac2206 100644
--- a/arch/arm/mm/fault.c
+++ b/arch/arm/mm/fault.c
@@ -108,7 +108,7 @@ void show_pte(struct mm_struct *mm, unsigned long addr)
 
                 pte = pte_offset_map(pmd, addr);
                 printk(", *pte=%08lx", pte_val(*pte));
-                printk(", *ppte=%08lx", pte_val(pte[-PTRS_PER_PTE]));
+                printk(", *ppte=%08lx", pte_val(pte[PTE_HWTABLE_PTRS]));
                 pte_unmap(pte);
         } while(0);
 
diff --git a/arch/arm/mm/idmap.c b/arch/arm/mm/idmap.c
new file mode 100644
index 000000000000..57299446f787
--- /dev/null
+++ b/arch/arm/mm/idmap.c
@@ -0,0 +1,67 @@
+#include <linux/kernel.h>
+
+#include <asm/cputype.h>
+#include <asm/pgalloc.h>
+#include <asm/pgtable.h>
+
+static void idmap_add_pmd(pgd_t *pgd, unsigned long addr, unsigned long end,
+        unsigned long prot)
+{
+        pmd_t *pmd = pmd_offset(pgd, addr);
+
+        addr = (addr & PMD_MASK) | prot;
+        pmd[0] = __pmd(addr);
+        addr += SECTION_SIZE;
+        pmd[1] = __pmd(addr);
+        flush_pmd_entry(pmd);
+}
+
+void identity_mapping_add(pgd_t *pgd, unsigned long addr, unsigned long end)
+{
+        unsigned long prot, next;
+
+        prot = PMD_TYPE_SECT | PMD_SECT_AP_WRITE;
+        if (cpu_architecture() <= CPU_ARCH_ARMv5TEJ && !cpu_is_xscale())
+                prot |= PMD_BIT4;
+
+        pgd += pgd_index(addr);
+        do {
+                next = pgd_addr_end(addr, end);
+                idmap_add_pmd(pgd, addr, next, prot);
+        } while (pgd++, addr = next, addr != end);
+}
+
+#ifdef CONFIG_SMP
+static void idmap_del_pmd(pgd_t *pgd, unsigned long addr, unsigned long end)
+{
+        pmd_t *pmd = pmd_offset(pgd, addr);
+        pmd_clear(pmd);
+}
+
+void identity_mapping_del(pgd_t *pgd, unsigned long addr, unsigned long end)
+{
+        unsigned long next;
+
+        pgd += pgd_index(addr);
+        do {
+                next = pgd_addr_end(addr, end);
+                idmap_del_pmd(pgd, addr, next);
+        } while (pgd++, addr = next, addr != end);
+}
+#endif
+
+/*
+ * In order to soft-boot, we need to insert a 1:1 mapping in place of
+ * the user-mode pages.  This will then ensure that we have predictable
+ * results when turning the mmu off
+ */
+void setup_mm_for_reboot(char mode)
+{
+        /*
+         * We need to access to user-mode page tables here. For kernel threads
+         * we don't have any user-mode mappings so we use the context that we
+         * "borrowed".
+         */
+        identity_mapping_add(current->active_mm->pgd, 0, TASK_SIZE);
+        local_flush_tlb_all();
+}
diff --git a/arch/arm/mm/mm.h b/arch/arm/mm/mm.h
index 6630620380a4..36960df5fb76 100644
--- a/arch/arm/mm/mm.h
+++ b/arch/arm/mm/mm.h
@@ -16,7 +16,7 @@ static inline pmd_t *pmd_off_k(unsigned long virt)
 }
 
 struct mem_type {
-        unsigned int prot_pte;
+        pteval_t prot_pte;
         unsigned int prot_l1;
         unsigned int prot_sect;
         unsigned int domain;
diff --git a/arch/arm/mm/mmu.c b/arch/arm/mm/mmu.c
index 79c01f540cbe..3c67e92f7d59 100644
--- a/arch/arm/mm/mmu.c
+++ b/arch/arm/mm/mmu.c
@@ -63,7 +63,7 @@ struct cachepolicy {
         const char      policy[16];
         unsigned int    cr_mask;
         unsigned int    pmd;
-        unsigned int    pte;
+        pteval_t        pte;
 };
 
 static struct cachepolicy cache_policies[] __initdata = {
@@ -191,7 +191,7 @@ void adjust_cr(unsigned long mask, unsigned long set)
 }
 #endif
 
-#define PROT_PTE_DEVICE         L_PTE_PRESENT|L_PTE_YOUNG|L_PTE_DIRTY|L_PTE_WRITE
+#define PROT_PTE_DEVICE         L_PTE_PRESENT|L_PTE_YOUNG|L_PTE_DIRTY|L_PTE_XN
 #define PROT_SECT_DEVICE        PMD_TYPE_SECT|PMD_SECT_AP_WRITE
 
 static struct mem_type mem_types[] = {
@@ -236,19 +236,18 @@ static struct mem_type mem_types[] = {
         },
         [MT_LOW_VECTORS] = {
                 .prot_pte  = L_PTE_PRESENT | L_PTE_YOUNG | L_PTE_DIRTY |
-                                L_PTE_EXEC,
+                                L_PTE_RDONLY,
                 .prot_l1   = PMD_TYPE_TABLE,
                 .domain    = DOMAIN_USER,
         },
         [MT_HIGH_VECTORS] = {
                 .prot_pte  = L_PTE_PRESENT | L_PTE_YOUNG | L_PTE_DIRTY |
-                                L_PTE_USER | L_PTE_EXEC,
+                                L_PTE_USER | L_PTE_RDONLY,
                 .prot_l1   = PMD_TYPE_TABLE,
                 .domain    = DOMAIN_USER,
         },
         [MT_MEMORY] = {
-                .prot_pte  = L_PTE_PRESENT | L_PTE_YOUNG | L_PTE_DIRTY |
-                                L_PTE_WRITE | L_PTE_EXEC,
+                .prot_pte  = L_PTE_PRESENT | L_PTE_YOUNG | L_PTE_DIRTY,
                 .prot_l1   = PMD_TYPE_TABLE,
                 .prot_sect = PMD_TYPE_SECT | PMD_SECT_AP_WRITE,
                 .domain    = DOMAIN_KERNEL,
@@ -259,21 +258,20 @@ static struct mem_type mem_types[] = {
         },
         [MT_MEMORY_NONCACHED] = {
                 .prot_pte  = L_PTE_PRESENT | L_PTE_YOUNG | L_PTE_DIRTY |
-                                L_PTE_WRITE | L_PTE_EXEC | L_PTE_MT_BUFFERABLE,
+                                L_PTE_MT_BUFFERABLE,
                 .prot_l1   = PMD_TYPE_TABLE,
                 .prot_sect = PMD_TYPE_SECT | PMD_SECT_AP_WRITE,
                 .domain    = DOMAIN_KERNEL,
         },
         [MT_MEMORY_DTCM] = {
                 .prot_pte  = L_PTE_PRESENT | L_PTE_YOUNG | L_PTE_DIRTY |
-                                L_PTE_WRITE,
+                                L_PTE_XN,
                 .prot_l1   = PMD_TYPE_TABLE,
                 .prot_sect = PMD_TYPE_SECT | PMD_SECT_XN,
                 .domain    = DOMAIN_KERNEL,
         },
         [MT_MEMORY_ITCM] = {
-                .prot_pte  = L_PTE_PRESENT | L_PTE_YOUNG | L_PTE_DIRTY |
-                                L_PTE_WRITE | L_PTE_EXEC,
+                .prot_pte  = L_PTE_PRESENT | L_PTE_YOUNG | L_PTE_DIRTY,
                 .prot_l1   = PMD_TYPE_TABLE,
                 .domain    = DOMAIN_KERNEL,
         },
@@ -480,7 +478,7 @@ static void __init build_mem_type_table(void)
 
         pgprot_user   = __pgprot(L_PTE_PRESENT | L_PTE_YOUNG | user_pgprot);
         pgprot_kernel = __pgprot(L_PTE_PRESENT | L_PTE_YOUNG |
-                                 L_PTE_DIRTY | L_PTE_WRITE | kern_pgprot);
+                                 L_PTE_DIRTY | kern_pgprot);
 
         mem_types[MT_LOW_VECTORS].prot_l1 |= ecc_mask;
         mem_types[MT_HIGH_VECTORS].prot_l1 |= ecc_mask;
@@ -536,7 +534,7 @@ static pte_t * __init early_pte_alloc(pmd_t *pmd, unsigned long addr, unsigned l
 {
         if (pmd_none(*pmd)) {
                 pte_t *pte = early_alloc(2 * PTRS_PER_PTE * sizeof(pte_t));
-                __pmd_populate(pmd, __pa(pte) | prot);
+                __pmd_populate(pmd, __pa(pte), prot);
         }
         BUG_ON(pmd_bad(*pmd));
         return pte_offset_kernel(pmd, addr);
@@ -554,7 +552,7 @@ static void __init alloc_init_pte(pmd_t *pmd, unsigned long addr,
 }
 
 static void __init alloc_init_section(pgd_t *pgd, unsigned long addr,
-                                      unsigned long end, unsigned long phys,
+                                      unsigned long end, phys_addr_t phys,
                                       const struct mem_type *type)
 {
         pmd_t *pmd = pmd_offset(pgd, addr);
@@ -589,7 +587,8 @@ static void __init alloc_init_section(pgd_t *pgd, unsigned long addr,
 static void __init create_36bit_mapping(struct map_desc *md,
                                         const struct mem_type *type)
 {
-        unsigned long phys, addr, length, end;
+        unsigned long addr, length, end;
+        phys_addr_t phys;
         pgd_t *pgd;
 
         addr = md->virtual;
@@ -1044,38 +1043,3 @@ void __init paging_init(struct machine_desc *mdesc)
         empty_zero_page = virt_to_page(zero_page);
         __flush_dcache_page(NULL, empty_zero_page);
 }
-
-/*
- * In order to soft-boot, we need to insert a 1:1 mapping in place of
- * the user-mode pages.  This will then ensure that we have predictable
- * results when turning the mmu off
- */
-void setup_mm_for_reboot(char mode)
-{
-        unsigned long base_pmdval;
-        pgd_t *pgd;
-        int i;
-
-        /*
-         * We need to access to user-mode page tables here. For kernel threads
-         * we don't have any user-mode mappings so we use the context that we
-         * "borrowed".
-         */
-        pgd = current->active_mm->pgd;
-
-        base_pmdval = PMD_SECT_AP_WRITE | PMD_SECT_AP_READ | PMD_TYPE_SECT;
-        if (cpu_architecture() <= CPU_ARCH_ARMv5TEJ && !cpu_is_xscale())
-                base_pmdval |= PMD_BIT4;
-
-        for (i = 0; i < FIRST_USER_PGD_NR + USER_PTRS_PER_PGD; i++, pgd++) {
-                unsigned long pmdval = (i << PGDIR_SHIFT) | base_pmdval;
-                pmd_t *pmd;
-
-                pmd = pmd_off(pgd, i << PGDIR_SHIFT);
-                pmd[0] = __pmd(pmdval);
-                pmd[1] = __pmd(pmdval + (1 << (PGDIR_SHIFT - 1)));
-                flush_pmd_entry(pmd);
-        }
-
-        local_flush_tlb_all();
-}
diff --git a/arch/arm/mm/pgd.c b/arch/arm/mm/pgd.c
index 69bbfc6645a6..93292a18cf77 100644
--- a/arch/arm/mm/pgd.c
+++ b/arch/arm/mm/pgd.c
@@ -17,12 +17,10 @@
 
 #include "mm.h"
 
-#define FIRST_KERNEL_PGD_NR     (FIRST_USER_PGD_NR + USER_PTRS_PER_PGD)
-
 /*
  * need to get a 16k page for level 1
  */
-pgd_t *get_pgd_slow(struct mm_struct *mm)
+pgd_t *pgd_alloc(struct mm_struct *mm)
 {
         pgd_t *new_pgd, *init_pgd;
         pmd_t *new_pmd, *init_pmd;
@@ -32,14 +30,14 @@ pgd_t *get_pgd_slow(struct mm_struct *mm)
         if (!new_pgd)
                 goto no_pgd;
 
-        memset(new_pgd, 0, FIRST_KERNEL_PGD_NR * sizeof(pgd_t));
+        memset(new_pgd, 0, USER_PTRS_PER_PGD * sizeof(pgd_t));
 
         /*
          * Copy over the kernel and IO PGD entries
          */
         init_pgd = pgd_offset_k(0);
-        memcpy(new_pgd + FIRST_KERNEL_PGD_NR, init_pgd + FIRST_KERNEL_PGD_NR,
-                       (PTRS_PER_PGD - FIRST_KERNEL_PGD_NR) * sizeof(pgd_t));
+        memcpy(new_pgd + USER_PTRS_PER_PGD, init_pgd + USER_PTRS_PER_PGD,
+                       (PTRS_PER_PGD - USER_PTRS_PER_PGD) * sizeof(pgd_t));
 
         clean_dcache_area(new_pgd, PTRS_PER_PGD * sizeof(pgd_t));
 
@@ -73,28 +71,29 @@ no_pgd:
         return NULL;
 }
 
-void free_pgd_slow(struct mm_struct *mm, pgd_t *pgd)
+void pgd_free(struct mm_struct *mm, pgd_t *pgd_base)
 {
+        pgd_t *pgd;
         pmd_t *pmd;
         pgtable_t pte;
 
-        if (!pgd)
+        if (!pgd_base)
                 return;
 
-        /* pgd is always present and good */
-        pmd = pmd_off(pgd, 0);
-        if (pmd_none(*pmd))
-                goto free;
-        if (pmd_bad(*pmd)) {
-                pmd_ERROR(*pmd);
-                pmd_clear(pmd);
-                goto free;
-        }
+        pgd = pgd_base + pgd_index(0);
+        if (pgd_none_or_clear_bad(pgd))
+                goto no_pgd;
+
+        pmd = pmd_offset(pgd, 0);
+        if (pmd_none_or_clear_bad(pmd))
+                goto no_pmd;
 
         pte = pmd_pgtable(*pmd);
         pmd_clear(pmd);
         pte_free(mm, pte);
+no_pmd:
+        pgd_clear(pgd);
         pmd_free(mm, pmd);
-free:
-        free_pages((unsigned long) pgd, 2);
+no_pgd:
+        free_pages((unsigned long) pgd_base, 2);
 }
diff --git a/arch/arm/mm/proc-macros.S b/arch/arm/mm/proc-macros.S
index f8f777df8d72..e32fa499194c 100644
--- a/arch/arm/mm/proc-macros.S
+++ b/arch/arm/mm/proc-macros.S
@@ -91,7 +91,7 @@
 #if L_PTE_SHARED != PTE_EXT_SHARED
 #error PTE shared bit mismatch
 #endif
-#if (L_PTE_EXEC+L_PTE_USER+L_PTE_WRITE+L_PTE_DIRTY+L_PTE_YOUNG+\
+#if (L_PTE_XN+L_PTE_USER+L_PTE_RDONLY+L_PTE_DIRTY+L_PTE_YOUNG+\
      L_PTE_FILE+L_PTE_PRESENT) > L_PTE_SHARED
 #error Invalid Linux PTE bit settings
 #endif
@@ -135,7 +135,7 @@
         .endm
 
         .macro  armv6_set_pte_ext pfx
-        str     r1, [r0], #-2048                @ linux version
+        str     r1, [r0], #2048                 @ linux version
 
         bic     r3, r1, #0x000003fc
         bic     r3, r3, #PTE_TYPE_MASK
@@ -146,9 +146,9 @@
         and     r2, r1, #L_PTE_MT_MASK
         ldr     r2, [ip, r2]
 
-        tst     r1, #L_PTE_WRITE
-        tstne   r1, #L_PTE_DIRTY
-        orreq   r3, r3, #PTE_EXT_APX
+        eor     r1, r1, #L_PTE_DIRTY
+        tst     r1, #L_PTE_DIRTY|L_PTE_RDONLY
+        orrne   r3, r3, #PTE_EXT_APX
 
         tst     r1, #L_PTE_USER
         orrne   r3, r3, #PTE_EXT_AP1
@@ -158,8 +158,8 @@
         bicne   r3, r3, #PTE_EXT_APX | PTE_EXT_AP0
 #endif
 
-        tst     r1, #L_PTE_EXEC
-        orreq   r3, r3, #PTE_EXT_XN
+        tst     r1, #L_PTE_XN
+        orrne   r3, r3, #PTE_EXT_XN
 
         orr     r3, r3, r2
 
@@ -187,9 +187,9 @@
  *  1111  0xff  r/w     r/w
  */
         .macro  armv3_set_pte_ext wc_disable=1
-        str     r1, [r0], #-2048                @ linux version
+        str     r1, [r0], #2048                 @ linux version
 
-        eor     r3, r1, #L_PTE_PRESENT | L_PTE_YOUNG | L_PTE_WRITE | L_PTE_DIRTY
+        eor     r3, r1, #L_PTE_PRESENT | L_PTE_YOUNG | L_PTE_DIRTY
 
         bic     r2, r1, #PTE_SMALL_AP_MASK      @ keep C, B bits
         bic     r2, r2, #PTE_TYPE_MASK
@@ -198,7 +198,7 @@
         tst     r3, #L_PTE_USER                 @ user?
         orrne   r2, r2, #PTE_SMALL_AP_URO_SRW
 
-        tst     r3, #L_PTE_WRITE | L_PTE_DIRTY  @ write and dirty?
+        tst     r3, #L_PTE_RDONLY | L_PTE_DIRTY @ write and dirty?
         orreq   r2, r2, #PTE_SMALL_AP_UNO_SRW
 
         tst     r3, #L_PTE_PRESENT | L_PTE_YOUNG        @ present and young?
@@ -210,7 +210,7 @@
         bicne   r2, r2, #PTE_BUFFERABLE
 #endif
         .endif
-        str     r2, [r0]                        @ hardware version
+        str     r2, [r0]                @ hardware version
         .endm
 
 
@@ -230,9 +230,9 @@
  *  1111  11    r/w     r/w
  */
         .macro  xscale_set_pte_ext_prologue
-        str     r1, [r0], #-2048                @ linux version
+        str     r1, [r0]                        @ linux version
 
-        eor     r3, r1, #L_PTE_PRESENT | L_PTE_YOUNG | L_PTE_WRITE | L_PTE_DIRTY
+        eor     r3, r1, #L_PTE_PRESENT | L_PTE_YOUNG | L_PTE_DIRTY
 
         bic     r2, r1, #PTE_SMALL_AP_MASK      @ keep C, B bits
         orr     r2, r2, #PTE_TYPE_EXT           @ extended page
@@ -240,7 +240,7 @@
         tst     r3, #L_PTE_USER                 @ user?
         orrne   r2, r2, #PTE_EXT_AP_URO_SRW     @ yes -> user r/o, system r/w
 
-        tst     r3, #L_PTE_WRITE | L_PTE_DIRTY  @ write and dirty?
+        tst     r3, #L_PTE_RDONLY | L_PTE_DIRTY @ write and dirty?
         orreq   r2, r2, #PTE_EXT_AP_UNO_SRW     @ yes -> user n/a, system r/w
                                                 @ combined with user -> user r/w
         .endm
@@ -249,7 +249,7 @@
         tst     r3, #L_PTE_PRESENT | L_PTE_YOUNG        @ present and young?
         movne   r2, #0                          @ no -> fault
 
-        str     r2, [r0]                        @ hardware version
+        str     r2, [r0, #2048]!                @ hardware version
         mov     ip, #0
         mcr     p15, 0, r0, c7, c10, 1          @ clean L1 D line
         mcr     p15, 0, ip, c7, c10, 4          @ data write barrier
diff --git a/arch/arm/mm/proc-v7.S b/arch/arm/mm/proc-v7.S
index 7401f4d7e676..b49fab21517c 100644
--- a/arch/arm/mm/proc-v7.S
+++ b/arch/arm/mm/proc-v7.S
@@ -124,15 +124,13 @@ ENDPROC(cpu_v7_switch_mm)
  *      Set a level 2 translation table entry.
  *
  *      - ptep  - pointer to level 2 translation table entry
- *                (hardware version is stored at -1024 bytes)
+ *                (hardware version is stored at +2048 bytes)
  *      - pte   - PTE value to store
  *      - ext   - value for extended PTE bits
  */
 ENTRY(cpu_v7_set_pte_ext)
 #ifdef CONFIG_MMU
- ARM(   str     r1, [r0], #-2048        )       @ linux version
- THUMB( str     r1, [r0]                )       @ linux version
- THUMB( sub     r0, r0, #2048           )
+        str     r1, [r0]                        @ linux version
 
         bic     r3, r1, #0x000003f0
         bic     r3, r3, #PTE_TYPE_MASK
@@ -142,9 +140,9 @@ ENTRY(cpu_v7_set_pte_ext)
         tst     r1, #1 << 4
         orrne   r3, r3, #PTE_EXT_TEX(1)
 
-        tst     r1, #L_PTE_WRITE
-        tstne   r1, #L_PTE_DIRTY
-        orreq   r3, r3, #PTE_EXT_APX
+        eor     r1, r1, #L_PTE_DIRTY
+        tst     r1, #L_PTE_RDONLY | L_PTE_DIRTY
+        orrne   r3, r3, #PTE_EXT_APX
 
         tst     r1, #L_PTE_USER
         orrne   r3, r3, #PTE_EXT_AP1
@@ -154,14 +152,14 @@ ENTRY(cpu_v7_set_pte_ext)
         bicne   r3, r3, #PTE_EXT_APX | PTE_EXT_AP0
 #endif
 
-        tst     r1, #L_PTE_EXEC
-        orreq   r3, r3, #PTE_EXT_XN
+        tst     r1, #L_PTE_XN
+        orrne   r3, r3, #PTE_EXT_XN
 
         tst     r1, #L_PTE_YOUNG
         tstne   r1, #L_PTE_PRESENT
         moveq   r3, #0
 
-        str     r3, [r0]
+        str     r3, [r0, #2048]!
         mcr     p15, 0, r0, c7, c10, 1          @ flush_pte
 #endif
         mov     pc, lr
diff --git a/arch/arm/mm/proc-xscale.S b/arch/arm/mm/proc-xscale.S
index 523408c0bb38..5a37c5e45c41 100644
--- a/arch/arm/mm/proc-xscale.S
+++ b/arch/arm/mm/proc-xscale.S
@@ -500,8 +500,8 @@ ENTRY(cpu_xscale_set_pte_ext)
         @
         @ Erratum 40: must set memory to write-through for user read-only pages
         @
-        and     ip, r1, #(L_PTE_MT_MASK | L_PTE_USER | L_PTE_WRITE) & ~(4 << 2)
-        teq     ip, #L_PTE_MT_WRITEBACK | L_PTE_USER
+        and     ip, r1, #(L_PTE_MT_MASK | L_PTE_USER | L_PTE_RDONLY) & ~(4 << 2)
+        teq     ip, #L_PTE_MT_WRITEBACK | L_PTE_USER | L_PTE_RDONLY
 
         moveq   r1, #L_PTE_MT_WRITETHROUGH
         and     r1, r1, #L_PTE_MT_MASK