X-Git-Url: http://ftp.safe.ca/?a=blobdiff_plain;f=arch%2Fx86%2Fmm%2Finit_64.c;h=151e5ba3441220a3f464dff6d3891dd4cb1d4367;hb=e7179853e7552ba6631e2cdd9f5c374383403b4b;hp=b10b7f17ea58d414576ded22b3c1b713fb06d6b6;hpb=22b45144f67dbaf0705992dc1462de2813fb83a1;p=safe%2Fjmp%2Flinux-2.6

diff --git a/arch/x86/mm/init_64.c b/arch/x86/mm/init_64.c
index b10b7f1..151e5ba 100644
--- a/arch/x86/mm/init_64.c
+++ b/arch/x86/mm/init_64.c
@@ -31,6 +31,7 @@
 #include <linux/nmi.h>
 
 #include <asm/processor.h>
+#include <asm/bios_ebda.h>
 #include <asm/system.h>
 #include <asm/uaccess.h>
 #include <asm/pgtable.h>
@@ -53,13 +54,14 @@
  * The direct mapping extends to max_pfn_mapped, so that we can directly access
  * apertures, ACPI and other tables without having to play with fixmaps.
  */
+unsigned long max_low_pfn_mapped;
 unsigned long max_pfn_mapped;
 
 static unsigned long dma_reserve __initdata;
 
 DEFINE_PER_CPU(struct mmu_gather, mmu_gathers);
 
-int direct_gbpages __meminitdata
+int direct_gbpages
 #ifdef CONFIG_DIRECT_GBPAGES
 				= 1
 #endif
@@ -85,46 +87,69 @@ early_param("gbpages", parse_direct_gbpages_on);
  * around without checking the pgd every time.
  */
 
-void show_mem(void)
-{
-	long i, total = 0, reserved = 0;
-	long shared = 0, cached = 0;
-	struct page *page;
-	pg_data_t *pgdat;
-
-	printk(KERN_INFO "Mem-info:\n");
-	show_free_areas();
-	for_each_online_pgdat(pgdat) {
-		for (i = 0; i < pgdat->node_spanned_pages; ++i) {
-			/*
-			 * This loop can take a while with 256 GB and
-			 * 4k pages so defer the NMI watchdog:
-			 */
-			if (unlikely(i % MAX_ORDER_NR_PAGES == 0))
-				touch_nmi_watchdog();
+int after_bootmem;
 
-			if (!pfn_valid(pgdat->node_start_pfn + i))
-				continue;
+pteval_t __supported_pte_mask __read_mostly = ~_PAGE_IOMAP;
+EXPORT_SYMBOL_GPL(__supported_pte_mask);
 
-			page = pfn_to_page(pgdat->node_start_pfn + i);
-			total++;
-			if (PageReserved(page))
-				reserved++;
-			else if (PageSwapCache(page))
-				cached++;
-			else if (page_count(page))
-				shared += page_count(page) - 1;
-		}
+static int do_not_nx __cpuinitdata;
+
+/*
+ * noexec=on|off
+ * Control non-executable mappings for 64-bit processes.
+ *
+ * on	Enable (default)
+ * off	Disable
+ */
+static int __init nonx_setup(char *str)
+{
+	if (!str)
+		return -EINVAL;
+	if (!strncmp(str, "on", 2)) {
+		__supported_pte_mask |= _PAGE_NX;
+		do_not_nx = 0;
+	} else if (!strncmp(str, "off", 3)) {
+		do_not_nx = 1;
+		__supported_pte_mask &= ~_PAGE_NX;
 	}
-	printk(KERN_INFO "%lu pages of RAM\n",		total);
-	printk(KERN_INFO "%lu reserved pages\n",	reserved);
-	printk(KERN_INFO "%lu pages shared\n",		shared);
-	printk(KERN_INFO "%lu pages swap cached\n",	cached);
+	return 0;
 }
+early_param("noexec", nonx_setup);
 
-int after_bootmem;
+void __cpuinit check_efer(void)
+{
+	unsigned long efer;
 
-static __init void *spp_getpage(void)
+	rdmsrl(MSR_EFER, efer);
+	if (!(efer & EFER_NX) || do_not_nx)
+		__supported_pte_mask &= ~_PAGE_NX;
+}
+
+int force_personality32;
+
+/*
+ * noexec32=on|off
+ * Control non executable heap for 32bit processes.
+ * To control the stack too use noexec=off
+ *
+ * on	PROT_READ does not imply PROT_EXEC for 32-bit processes (default)
+ * off	PROT_READ implies PROT_EXEC
+ */
+static int __init nonx32_setup(char *str)
+{
+	if (!strcmp(str, "on"))
+		force_personality32 &= ~READ_IMPLIES_EXEC;
+	else if (!strcmp(str, "off"))
+		force_personality32 |= READ_IMPLIES_EXEC;
+	return 1;
+}
+__setup("noexec32=", nonx32_setup);
+
+/*
+ * NOTE: This function is marked __ref because it calls __init function
+ * (alloc_bootmem_pages). It's safe to do it ONLY when after_bootmem == 0.
+ */
+static __ref void *spp_getpage(void)
 {
 	void *ptr;
 
@@ -171,9 +196,6 @@ set_pte_vaddr_pud(pud_t *pud_page, unsigned long vaddr, pte_t new_pte)
 	}
 
 	pte = pte_offset_kernel(pmd, vaddr);
-	if (!pte_none(*pte) && pte_val(new_pte) &&
-	    pte_val(*pte) != (pte_val(new_pte) & __supported_pte_mask))
-		pte_ERROR(*pte);
 	set_pte(pte, new_pte);
 
 	/*
@@ -202,6 +224,46 @@ set_pte_vaddr(unsigned long vaddr, pte_t pteval)
 }
 
 /*
+ * Create large page table mappings for a range of physical addresses.
+ */
+static void __init __init_extra_mapping(unsigned long phys, unsigned long size,
+						pgprot_t prot)
+{
+	pgd_t *pgd;
+	pud_t *pud;
+	pmd_t *pmd;
+
+	BUG_ON((phys & ~PMD_MASK) || (size & ~PMD_MASK));
+	for (; size; phys += PMD_SIZE, size -= PMD_SIZE) {
+		pgd = pgd_offset_k((unsigned long)__va(phys));
+		if (pgd_none(*pgd)) {
+			pud = (pud_t *) spp_getpage();
+			set_pgd(pgd, __pgd(__pa(pud) | _KERNPG_TABLE |
+						_PAGE_USER));
+		}
+		pud = pud_offset(pgd, (unsigned long)__va(phys));
+		if (pud_none(*pud)) {
+			pmd = (pmd_t *) spp_getpage();
+			set_pud(pud, __pud(__pa(pmd) | _KERNPG_TABLE |
+						_PAGE_USER));
+		}
+		pmd = pmd_offset(pud, phys);
+		BUG_ON(!pmd_none(*pmd));
+		set_pmd(pmd, __pmd(phys | pgprot_val(prot)));
+	}
+}
+
+void __init init_extra_mapping_wb(unsigned long phys, unsigned long size)
+{
+	__init_extra_mapping(phys, size, PAGE_KERNEL_LARGE);
+}
+
+void __init init_extra_mapping_uc(unsigned long phys, unsigned long size)
+{
+	__init_extra_mapping(phys, size, PAGE_KERNEL_LARGE_NOCACHE);
+}
+
+/*
  * The head.S code sets up the kernel high mapping:
  *
  *   from __START_KERNEL_map to __START_KERNEL_map + size (== _end-_text)
@@ -217,7 +279,7 @@ set_pte_vaddr(unsigned long vaddr, pte_t pteval)
 void __init cleanup_highmap(void)
 {
 	unsigned long vaddr = __START_KERNEL_map;
-	unsigned long end = round_up((unsigned long)_end, PMD_SIZE) - 1;
+	unsigned long end = roundup((unsigned long)_end, PMD_SIZE) - 1;
 	pmd_t *pmd = level2_kernel_pgt;
 	pmd_t *last_pmd = pmd + PTRS_PER_PMD;
 
@@ -233,7 +295,7 @@ static unsigned long __initdata table_start;
 static unsigned long __meminitdata table_end;
 static unsigned long __meminitdata table_top;
 
-static __meminit void *alloc_low_page(unsigned long *phys)
+static __ref void *alloc_low_page(unsigned long *phys)
 {
 	unsigned long pfn = table_end++;
 	void *adr;
@@ -248,13 +310,13 @@ static __meminit void *alloc_low_page(unsigned long *phys)
 	if (pfn >= table_top)
 		panic("alloc_low_page: ran out of memory");
 
-	adr = early_ioremap(pfn * PAGE_SIZE, PAGE_SIZE);
+	adr = early_memremap(pfn * PAGE_SIZE, PAGE_SIZE);
 	memset(adr, 0, PAGE_SIZE);
 	*phys  = pfn * PAGE_SIZE;
 	return adr;
 }
 
-static __meminit void unmap_low_page(void *adr)
+static __ref void unmap_low_page(void *adr)
 {
 	if (after_bootmem)
 		return;
@@ -262,11 +324,14 @@ static __meminit void unmap_low_page(void *adr)
 	early_iounmap(adr, PAGE_SIZE);
 }
 
-static void __meminit
-phys_pte_init(pte_t *pte_page, unsigned long addr, unsigned long end)
+static unsigned long __meminit
+phys_pte_init(pte_t *pte_page, unsigned long addr, unsigned long end,
+	      pgprot_t prot)
 {
 	unsigned pages = 0;
+	unsigned long last_map_addr = end;
 	int i;
+
 	pte_t *pte = pte_page + pte_index(addr);
 
 	for(i = pte_index(addr); i < PTRS_PER_PTE; i++, addr += PAGE_SIZE, pte++) {
@@ -279,30 +344,45 @@ phys_pte_init(pte_t *pte_page, unsigned long addr, unsigned long end)
 			break;
 		}
 
-		if (pte_val(*pte))
+		/*
+		 * We will re-use the existing mapping.
+		 * Xen for example has some special requirements, like mapping
+		 * pagetable pages as RO. So assume someone who pre-setup
+		 * these mappings are more intelligent.
+		 */
+		if (pte_val(*pte)) {
+			pages++;
 			continue;
+		}
 
 		if (0)
 			printk("   pte=%p addr=%lx pte=%016lx\n",
 			       pte, addr, pfn_pte(addr >> PAGE_SHIFT, PAGE_KERNEL).pte);
-		set_pte(pte, pfn_pte(addr >> PAGE_SHIFT, PAGE_KERNEL));
 		pages++;
+		set_pte(pte, pfn_pte(addr >> PAGE_SHIFT, prot));
+		last_map_addr = (addr & PAGE_MASK) + PAGE_SIZE;
 	}
+
 	update_page_count(PG_LEVEL_4K, pages);
+
+	return last_map_addr;
 }
 
-static void __meminit
-phys_pte_update(pmd_t *pmd, unsigned long address, unsigned long end)
+static unsigned long __meminit
+phys_pte_update(pmd_t *pmd, unsigned long address, unsigned long end,
+		pgprot_t prot)
 {
 	pte_t *pte = (pte_t *)pmd_page_vaddr(*pmd);
 
-	phys_pte_init(pte, address, end);
+	return phys_pte_init(pte, address, end, prot);
 }
 
 static unsigned long __meminit
-phys_pmd_init(pmd_t *pmd_page, unsigned long address, unsigned long end)
+phys_pmd_init(pmd_t *pmd_page, unsigned long address, unsigned long end,
+	      unsigned long page_size_mask, pgprot_t prot)
 {
 	unsigned long pages = 0;
+	unsigned long last_map_addr = end;
 
 	int i = pmd_index(address);
 
@@ -310,6 +390,7 @@ phys_pmd_init(pmd_t *pmd_page, unsigned long address, unsigned long end)
 		unsigned long pte_phys;
 		pmd_t *pmd = pmd_page + pmd_index(address);
 		pte_t *pte;
+		pgprot_t new_prot = prot;
 
 		if (address >= end) {
 			if (!after_bootmem) {
@@ -320,43 +401,70 @@ phys_pmd_init(pmd_t *pmd_page, unsigned long address, unsigned long end)
 		}
 
 		if (pmd_val(*pmd)) {
-			if (!pmd_large(*pmd))
-				phys_pte_update(pmd, address, end);
-			continue;
+			if (!pmd_large(*pmd)) {
+				spin_lock(&init_mm.page_table_lock);
+				last_map_addr = phys_pte_update(pmd, address,
+								end, prot);
+				spin_unlock(&init_mm.page_table_lock);
+				continue;
+			}
+			/*
+			 * If we are ok with PG_LEVEL_2M mapping, then we will
+			 * use the existing mapping,
+			 *
+			 * Otherwise, we will split the large page mapping but
+			 * use the same existing protection bits except for
+			 * large page, so that we don't violate Intel's TLB
+			 * Application note (317080) which says, while changing
+			 * the page sizes, new and old translations should
+			 * not differ with respect to page frame and
+			 * attributes.
+			 */
+			if (page_size_mask & (1 << PG_LEVEL_2M)) {
+				pages++;
+				continue;
+			}
+			new_prot = pte_pgprot(pte_clrhuge(*(pte_t *)pmd));
 		}
 
-		if (cpu_has_pse) {
+		if (page_size_mask & (1<<PG_LEVEL_2M)) {
 			pages++;
+			spin_lock(&init_mm.page_table_lock);
 			set_pte((pte_t *)pmd,
-				pfn_pte(address >> PAGE_SHIFT, PAGE_KERNEL_LARGE));
+				pfn_pte(address >> PAGE_SHIFT,
+					__pgprot(pgprot_val(prot) | _PAGE_PSE)));
+			spin_unlock(&init_mm.page_table_lock);
+			last_map_addr = (address & PMD_MASK) + PMD_SIZE;
 			continue;
 		}
 
 		pte = alloc_low_page(&pte_phys);
-		phys_pte_init(pte, address, end);
+		last_map_addr = phys_pte_init(pte, address, end, new_prot);
 		unmap_low_page(pte);
 
+		spin_lock(&init_mm.page_table_lock);
 		pmd_populate_kernel(&init_mm, pmd, __va(pte_phys));
+		spin_unlock(&init_mm.page_table_lock);
 	}
 	update_page_count(PG_LEVEL_2M, pages);
-	return address;
+	return last_map_addr;
 }
 
 static unsigned long __meminit
-phys_pmd_update(pud_t *pud, unsigned long address, unsigned long end)
+phys_pmd_update(pud_t *pud, unsigned long address, unsigned long end,
+		unsigned long page_size_mask, pgprot_t prot)
 {
 	pmd_t *pmd = pmd_offset(pud, 0);
 	unsigned long last_map_addr;
 
-	spin_lock(&init_mm.page_table_lock);
-	last_map_addr = phys_pmd_init(pmd, address, end);
-	spin_unlock(&init_mm.page_table_lock);
+	last_map_addr = phys_pmd_init(pmd, address, end, page_size_mask, prot);
 	__flush_tlb_all();
 	return last_map_addr;
 }
 
 static unsigned long __meminit
-phys_pud_init(pud_t *pud_page, unsigned long addr, unsigned long end)
+phys_pud_init(pud_t *pud_page, unsigned long addr, unsigned long end,
+			 unsigned long page_size_mask)
 {
 	unsigned long pages = 0;
 	unsigned long last_map_addr = end;
@@ -366,6 +474,7 @@ phys_pud_init(pud_t *pud_page, unsigned long addr, unsigned long end)
 		unsigned long pmd_phys;
 		pud_t *pud = pud_page + pud_index(addr);
 		pmd_t *pmd;
+		pgprot_t prot = PAGE_KERNEL;
 
 		if (addr >= end)
 			break;
@@ -377,66 +486,116 @@ phys_pud_init(pud_t *pud_page, unsigned long addr, unsigned long end)
 		}
 
 		if (pud_val(*pud)) {
-			if (!pud_large(*pud))
-				last_map_addr = phys_pmd_update(pud, addr, end);
-			continue;
+			if (!pud_large(*pud)) {
+				last_map_addr = phys_pmd_update(pud, addr, end,
+							 page_size_mask, prot);
+				continue;
+			}
+			/*
+			 * If we are ok with PG_LEVEL_1G mapping, then we will
+			 * use the existing mapping.
+			 *
+			 * Otherwise, we will split the gbpage mapping but use
+			 * the same existing protection  bits except for large
+			 * page, so that we don't violate Intel's TLB
+			 * Application note (317080) which says, while changing
+			 * the page sizes, new and old translations should
+			 * not differ with respect to page frame and
+			 * attributes.
+			 */
+			if (page_size_mask & (1 << PG_LEVEL_1G)) {
+				pages++;
+				continue;
+			}
+			prot = pte_pgprot(pte_clrhuge(*(pte_t *)pud));
 		}
 
-		if (direct_gbpages) {
+		if (page_size_mask & (1<<PG_LEVEL_1G)) {
 			pages++;
+			spin_lock(&init_mm.page_table_lock);
 			set_pte((pte_t *)pud,
 				pfn_pte(addr >> PAGE_SHIFT, PAGE_KERNEL_LARGE));
+			spin_unlock(&init_mm.page_table_lock);
 			last_map_addr = (addr & PUD_MASK) + PUD_SIZE;
 			continue;
 		}
 
 		pmd = alloc_low_page(&pmd_phys);
+		last_map_addr = phys_pmd_init(pmd, addr, end, page_size_mask,
+					      prot);
+		unmap_low_page(pmd);
 
 		spin_lock(&init_mm.page_table_lock);
-		last_map_addr = phys_pmd_init(pmd, addr, end);
-		unmap_low_page(pmd);
 		pud_populate(&init_mm, pud, __va(pmd_phys));
 		spin_unlock(&init_mm.page_table_lock);
-
 	}
 	__flush_tlb_all();
+
 	update_page_count(PG_LEVEL_1G, pages);
 
 	return last_map_addr;
 }
 
 static unsigned long __meminit
-phys_pud_update(pgd_t *pgd, unsigned long addr, unsigned long end)
+phys_pud_update(pgd_t *pgd, unsigned long addr, unsigned long end,
+		 unsigned long page_size_mask)
 {
 	pud_t *pud;
 
 	pud = (pud_t *)pgd_page_vaddr(*pgd);
 
-	return phys_pud_init(pud, addr, end);
+	return phys_pud_init(pud, addr, end, page_size_mask);
 }
 
-static void __init find_early_table_space(unsigned long end)
+static void __init find_early_table_space(unsigned long end, int use_pse,
+					  int use_gbpages)
 {
-	unsigned long puds, tables, start;
+	unsigned long puds, pmds, ptes, tables, start;
 
 	puds = (end + PUD_SIZE - 1) >> PUD_SHIFT;
-	tables = round_up(puds * sizeof(pud_t), PAGE_SIZE);
-	if (!direct_gbpages) {
-		unsigned long pmds = (end + PMD_SIZE - 1) >> PMD_SHIFT;
-		tables += round_up(pmds * sizeof(pmd_t), PAGE_SIZE);
-	}
-	if (!cpu_has_pse) {
-		unsigned long ptes = (end + PAGE_SIZE - 1) >> PAGE_SHIFT;
-		tables += round_up(ptes * sizeof(pte_t), PAGE_SIZE);
-	}
+	tables = roundup(puds * sizeof(pud_t), PAGE_SIZE);
+
+	if (use_gbpages) {
+		unsigned long extra;
+
+		extra = end - ((end>>PUD_SHIFT) << PUD_SHIFT);
+		pmds = (extra + PMD_SIZE - 1) >> PMD_SHIFT;
+	} else
+		pmds = (end + PMD_SIZE - 1) >> PMD_SHIFT;
+
+	tables += roundup(pmds * sizeof(pmd_t), PAGE_SIZE);
+
+	if (use_pse) {
+		unsigned long extra;
+
+		extra = end - ((end>>PMD_SHIFT) << PMD_SHIFT);
+#ifdef CONFIG_X86_32
+		extra += PMD_SIZE;
+#endif
+		ptes = (extra + PAGE_SIZE - 1) >> PAGE_SHIFT;
+	} else
+		ptes = (end + PAGE_SIZE - 1) >> PAGE_SHIFT;
+
+	tables += roundup(ptes * sizeof(pte_t), PAGE_SIZE);
+
+#ifdef CONFIG_X86_32
+	/* for fixmap */
+	tables += roundup(__end_of_fixed_addresses * sizeof(pte_t), PAGE_SIZE);
+#endif
 
 	/*
 	 * RED-PEN putting page tables only on node 0 could
 	 * cause a hotspot and fill up ZONE_DMA. The page tables
 	 * need roughly 0.5KB per GB.
 	 */
+#ifdef CONFIG_X86_32
+	start = 0x7000;
+	table_start = find_e820_area(start, max_pfn_mapped<<PAGE_SHIFT,
+					tables, PAGE_SIZE);
+#else /* CONFIG_X86_64 */
 	start = 0x8000;
 	table_start = find_e820_area(start, end, tables, PAGE_SIZE);
+#endif
 	if (table_start == -1UL)
 		panic("Cannot find space for the kernel page tables");
 
@@ -456,181 +615,210 @@ static void __init init_gbpages(void)
 		direct_gbpages = 0;
 }
 
-#ifdef CONFIG_MEMTEST
-
-static void __init memtest(unsigned long start_phys, unsigned long size,
-				 unsigned pattern)
+static unsigned long __meminit kernel_physical_mapping_init(unsigned long start,
+						unsigned long end,
+						unsigned long page_size_mask)
 {
-	unsigned long i;
-	unsigned long *start;
-	unsigned long start_bad;
-	unsigned long last_bad;
-	unsigned long val;
-	unsigned long start_phys_aligned;
-	unsigned long count;
-	unsigned long incr;
-
-	switch (pattern) {
-	case 0:
-		val = 0UL;
-		break;
-	case 1:
-		val = -1UL;
-		break;
-	case 2:
-		val = 0x5555555555555555UL;
-		break;
-	case 3:
-		val = 0xaaaaaaaaaaaaaaaaUL;
-		break;
-	default:
-		return;
-	}
-
-	incr = sizeof(unsigned long);
-	start_phys_aligned = ALIGN(start_phys, incr);
-	count = (size - (start_phys_aligned - start_phys))/incr;
-	start = __va(start_phys_aligned);
-	start_bad = 0;
-	last_bad = 0;
-
-	for (i = 0; i < count; i++)
-		start[i] = val;
-	for (i = 0; i < count; i++, start++, start_phys_aligned += incr) {
-		if (*start != val) {
-			if (start_phys_aligned == last_bad + incr) {
-				last_bad += incr;
-			} else {
-				if (start_bad) {
-					printk(KERN_CONT "\n  %016lx bad mem addr %016lx - %016lx reserved",
-						val, start_bad, last_bad + incr);
-					reserve_early(start_bad, last_bad - start_bad, "BAD RAM");
-				}
-				start_bad = last_bad = start_phys_aligned;
-			}
-		}
-	}
-	if (start_bad) {
-		printk(KERN_CONT "\n  %016lx bad mem addr %016lx - %016lx reserved",
-			val, start_bad, last_bad + incr);
-		reserve_early(start_bad, last_bad - start_bad, "BAD RAM");
-	}
 
-}
+	unsigned long next, last_map_addr = end;
 
-/* default is disabled */
-static int memtest_pattern __initdata;
+	start = (unsigned long)__va(start);
+	end = (unsigned long)__va(end);
 
-static int __init parse_memtest(char *arg)
-{
-	if (arg)
-		memtest_pattern = simple_strtoul(arg, NULL, 0);
-	return 0;
-}
+	for (; start < end; start = next) {
+		pgd_t *pgd = pgd_offset_k(start);
+		unsigned long pud_phys;
+		pud_t *pud;
 
-early_param("memtest", parse_memtest);
+		next = (start + PGDIR_SIZE) & PGDIR_MASK;
+		if (next > end)
+			next = end;
 
-static void __init early_memtest(unsigned long start, unsigned long end)
-{
-	u64 t_start, t_size;
-	unsigned pattern;
+		if (pgd_val(*pgd)) {
+			last_map_addr = phys_pud_update(pgd, __pa(start),
+						 __pa(end), page_size_mask);
+			continue;
+		}
 
-	if (!memtest_pattern)
-		return;
+		pud = alloc_low_page(&pud_phys);
+		last_map_addr = phys_pud_init(pud, __pa(start), __pa(next),
+						 page_size_mask);
+		unmap_low_page(pud);
 
-	printk(KERN_INFO "early_memtest: pattern num %d", memtest_pattern);
-	for (pattern = 0; pattern < memtest_pattern; pattern++) {
-		t_start = start;
-		t_size = 0;
-		while (t_start < end) {
-			t_start = find_e820_area_size(t_start, &t_size, 1);
+		spin_lock(&init_mm.page_table_lock);
+		pgd_populate(&init_mm, pgd, __va(pud_phys));
+		spin_unlock(&init_mm.page_table_lock);
+	}
+	__flush_tlb_all();
 
-			/* done ? */
-			if (t_start >= end)
-				break;
-			if (t_start + t_size > end)
-				t_size = end - t_start;
+	return last_map_addr;
+}
 
-			printk(KERN_CONT "\n  %016llx - %016llx pattern %d",
-				(unsigned long long)t_start,
-				(unsigned long long)t_start + t_size, pattern);
+struct map_range {
+	unsigned long start;
+	unsigned long end;
+	unsigned page_size_mask;
+};
 
-			memtest(t_start, t_size, pattern);
+#define NR_RANGE_MR 5
 
-			t_start += t_size;
-		}
-	}
-	printk(KERN_CONT "\n");
-}
-#else
-static void __init early_memtest(unsigned long start, unsigned long end)
+static int save_mr(struct map_range *mr, int nr_range,
+		   unsigned long start_pfn, unsigned long end_pfn,
+		   unsigned long page_size_mask)
 {
+	if (start_pfn < end_pfn) {
+		if (nr_range >= NR_RANGE_MR)
+			panic("run out of range for init_memory_mapping\n");
+		mr[nr_range].start = start_pfn<<PAGE_SHIFT;
+		mr[nr_range].end   = end_pfn<<PAGE_SHIFT;
+		mr[nr_range].page_size_mask = page_size_mask;
+		nr_range++;
+	}
+
+	return nr_range;
 }
-#endif
 
 /*
  * Setup the direct mapping of the physical memory at PAGE_OFFSET.
  * This runs before bootmem is initialized and gets pages directly from
  * the physical memory. To access them they are temporarily mapped.
  */
-unsigned long __init_refok init_memory_mapping(unsigned long start, unsigned long end)
+unsigned long __init_refok init_memory_mapping(unsigned long start,
+					       unsigned long end)
 {
-	unsigned long next, last_map_addr = end;
-	unsigned long start_phys = start, end_phys = end;
+	unsigned long last_map_addr = 0;
+	unsigned long page_size_mask = 0;
+	unsigned long start_pfn, end_pfn;
+	unsigned long pos;
+
+	struct map_range mr[NR_RANGE_MR];
+	int nr_range, i;
+	int use_pse, use_gbpages;
 
-	printk(KERN_INFO "init_memory_mapping\n");
+	printk(KERN_INFO "init_memory_mapping: %016lx-%016lx\n", start, end);
 
+	if (!after_bootmem)
+		init_gbpages();
+
+#ifdef CONFIG_DEBUG_PAGEALLOC
 	/*
-	 * Find space for the kernel direct mapping tables.
-	 *
-	 * Later we should allocate these tables in the local node of the
-	 * memory mapped. Unfortunately this is done currently before the
-	 * nodes are discovered.
+	 * For CONFIG_DEBUG_PAGEALLOC, identity mapping will use small pages.
+	 * This will simplify cpa(), which otherwise needs to support splitting
+	 * large pages into small in interrupt context, etc.
 	 */
-	if (!after_bootmem) {
-		init_gbpages();
-		find_early_table_space(end);
+	use_pse = use_gbpages = 0;
+#else
+	use_pse = cpu_has_pse;
+	use_gbpages = direct_gbpages;
+#endif
+
+	if (use_gbpages)
+		page_size_mask |= 1 << PG_LEVEL_1G;
+	if (use_pse)
+		page_size_mask |= 1 << PG_LEVEL_2M;
+
+	memset(mr, 0, sizeof(mr));
+	nr_range = 0;
+
+	/* head if not big page alignment ? */
+	start_pfn = start >> PAGE_SHIFT;
+	pos = start_pfn << PAGE_SHIFT;
+	end_pfn = ((pos + (PMD_SIZE - 1)) >> PMD_SHIFT)
+			<< (PMD_SHIFT - PAGE_SHIFT);
+	if (end_pfn > (end >> PAGE_SHIFT))
+		end_pfn = end >> PAGE_SHIFT;
+	if (start_pfn < end_pfn) {
+		nr_range = save_mr(mr, nr_range, start_pfn, end_pfn, 0);
+		pos = end_pfn << PAGE_SHIFT;
 	}
 
-	start = (unsigned long)__va(start);
-	end = (unsigned long)__va(end);
+	/* big page (2M) range */
+	start_pfn = ((pos + (PMD_SIZE - 1))>>PMD_SHIFT)
+			 << (PMD_SHIFT - PAGE_SHIFT);
+	end_pfn = ((pos + (PUD_SIZE - 1))>>PUD_SHIFT)
+			 << (PUD_SHIFT - PAGE_SHIFT);
+	if (end_pfn > ((end>>PMD_SHIFT)<<(PMD_SHIFT - PAGE_SHIFT)))
+		end_pfn = ((end>>PMD_SHIFT)<<(PMD_SHIFT - PAGE_SHIFT));
+	if (start_pfn < end_pfn) {
+		nr_range = save_mr(mr, nr_range, start_pfn, end_pfn,
+				page_size_mask & (1<<PG_LEVEL_2M));
+		pos = end_pfn << PAGE_SHIFT;
+	}
 
-	for (; start < end; start = next) {
-		pgd_t *pgd = pgd_offset_k(start);
-		unsigned long pud_phys;
-		pud_t *pud;
+	/* big page (1G) range */
+	start_pfn = ((pos + (PUD_SIZE - 1))>>PUD_SHIFT)
+			 << (PUD_SHIFT - PAGE_SHIFT);
+	end_pfn = (end >> PUD_SHIFT) << (PUD_SHIFT - PAGE_SHIFT);
+	if (start_pfn < end_pfn) {
+		nr_range = save_mr(mr, nr_range, start_pfn, end_pfn,
+				page_size_mask &
+				 ((1<<PG_LEVEL_2M)|(1<<PG_LEVEL_1G)));
+		pos = end_pfn << PAGE_SHIFT;
+	}
 
-		next = start + PGDIR_SIZE;
-		if (next > end)
-			next = end;
+	/* tail is not big page (1G) alignment */
+	start_pfn = ((pos + (PMD_SIZE - 1))>>PMD_SHIFT)
+			 << (PMD_SHIFT - PAGE_SHIFT);
+	end_pfn = (end >> PMD_SHIFT) << (PMD_SHIFT - PAGE_SHIFT);
+	if (start_pfn < end_pfn) {
+		nr_range = save_mr(mr, nr_range, start_pfn, end_pfn,
+				page_size_mask & (1<<PG_LEVEL_2M));
+		pos = end_pfn << PAGE_SHIFT;
+	}
 
-		if (pgd_val(*pgd)) {
-			last_map_addr = phys_pud_update(pgd, __pa(start), __pa(end));
+	/* tail is not big page (2M) alignment */
+	start_pfn = pos>>PAGE_SHIFT;
+	end_pfn = end>>PAGE_SHIFT;
+	nr_range = save_mr(mr, nr_range, start_pfn, end_pfn, 0);
+
+	/* try to merge same page size and continuous */
+	for (i = 0; nr_range > 1 && i < nr_range - 1; i++) {
+		unsigned long old_start;
+		if (mr[i].end != mr[i+1].start ||
+		    mr[i].page_size_mask != mr[i+1].page_size_mask)
 			continue;
-		}
+		/* move it */
+		old_start = mr[i].start;
+		memmove(&mr[i], &mr[i+1],
+			(nr_range - 1 - i) * sizeof(struct map_range));
+		mr[i--].start = old_start;
+		nr_range--;
+	}
 
-		if (after_bootmem)
-			pud = pud_offset(pgd, start & PGDIR_MASK);
-		else
-			pud = alloc_low_page(&pud_phys);
+	for (i = 0; i < nr_range; i++)
+		printk(KERN_DEBUG " %010lx - %010lx page %s\n",
+				mr[i].start, mr[i].end,
+			(mr[i].page_size_mask & (1<<PG_LEVEL_1G))?"1G":(
+			 (mr[i].page_size_mask & (1<<PG_LEVEL_2M))?"2M":"4k"));
 
-		last_map_addr = phys_pud_init(pud, __pa(start), __pa(next));
-		unmap_low_page(pud);
-		if (!after_bootmem)
-			pgd_populate(&init_mm, pgd_offset_k(start),
-				     __va(pud_phys));
-	}
+	/*
+	 * Find space for the kernel direct mapping tables.
+	 *
+	 * Later we should allocate these tables in the local node of the
+	 * memory mapped. Unfortunately this is done currently before the
+	 * nodes are discovered.
+	 */
+	if (!after_bootmem)
+		find_early_table_space(end, use_pse, use_gbpages);
+
+	for (i = 0; i < nr_range; i++)
+		last_map_addr = kernel_physical_mapping_init(
+					mr[i].start, mr[i].end,
+					mr[i].page_size_mask);
 
 	if (!after_bootmem)
 		mmu_cr4_features = read_cr4();
 	__flush_tlb_all();
 
-	if (!after_bootmem)
+	if (!after_bootmem && table_end > table_start)
 		reserve_early(table_start << PAGE_SHIFT,
 				 table_end << PAGE_SHIFT, "PGTABLE");
 
+	printk(KERN_INFO "last_map_addr: %lx end: %lx\n",
+			 last_map_addr, end);
+
 	if (!after_bootmem)
-		early_memtest(start_phys, end_phys);
+		early_memtest(start, end);
 
 	return last_map_addr >> PAGE_SHIFT;
 }
@@ -685,12 +873,12 @@ int arch_add_memory(int nid, u64 start, u64 size)
 	unsigned long nr_pages = size >> PAGE_SHIFT;
 	int ret;
 
-	last_mapped_pfn = init_memory_mapping(start, start + size-1);
+	last_mapped_pfn = init_memory_mapping(start, start + size);
 	if (last_mapped_pfn > max_pfn_mapped)
 		max_pfn_mapped = last_mapped_pfn;
 
-	ret = __add_pages(zone, start_pfn, nr_pages);
-	WARN_ON(1);
+	ret = __add_pages(nid, zone, start_pfn, nr_pages);
+	WARN_ON_ONCE(ret);
 
 	return ret;
 }
@@ -706,32 +894,13 @@ EXPORT_SYMBOL_GPL(memory_add_physaddr_to_nid);
 
 #endif /* CONFIG_MEMORY_HOTPLUG */
 
-/*
- * devmem_is_allowed() checks to see if /dev/mem access to a certain address
- * is valid. The argument is a physical page number.
- *
- *
- * On x86, access has to be given to the first megabyte of ram because that area
- * contains bios code and data regions used by X and dosemu and similar apps.
- * Access has to be given to non-kernel-ram areas as well, these contain the PCI
- * mmio resources as well as potential bios/acpi data regions.
- */
-int devmem_is_allowed(unsigned long pagenr)
-{
-	if (pagenr <= 256)
-		return 1;
-	if (!page_is_ram(pagenr))
-		return 1;
-	return 0;
-}
-
-
 static struct kcore_list kcore_mem, kcore_vmalloc, kcore_kernel,
 			 kcore_modules, kcore_vsyscall;
 
 void __init mem_init(void)
 {
 	long codesize, reservedpages, datasize, initsize;
+	unsigned long absent_pages;
 
 	pci_iommu_alloc();
 
@@ -745,8 +914,9 @@ void __init mem_init(void)
 #else
 	totalram_pages = free_all_bootmem();
 #endif
-	reservedpages = max_pfn - totalram_pages -
-					absent_pages_in_range(0, max_pfn);
+
+	absent_pages = absent_pages_in_range(0, max_pfn);
+	reservedpages = max_pfn - totalram_pages - absent_pages;
 	after_bootmem = 1;
 
 	codesize =  (unsigned long) &_etext - (unsigned long) &_text;
@@ -763,52 +933,14 @@ void __init mem_init(void)
 				 VSYSCALL_END - VSYSCALL_START);
 
 	printk(KERN_INFO "Memory: %luk/%luk available (%ldk kernel code, "
-				"%ldk reserved, %ldk data, %ldk init)\n",
+			 "%ldk absent, %ldk reserved, %ldk data, %ldk init)\n",
 		(unsigned long) nr_free_pages() << (PAGE_SHIFT-10),
 		max_pfn << (PAGE_SHIFT-10),
 		codesize >> 10,
+		absent_pages << (PAGE_SHIFT-10),
 		reservedpages << (PAGE_SHIFT-10),
 		datasize >> 10,
 		initsize >> 10);
-
-	cpa_init();
-}
-
-void free_init_pages(char *what, unsigned long begin, unsigned long end)
-{
-	unsigned long addr = begin;
-
-	if (addr >= end)
-		return;
-
-	/*
-	 * If debugging page accesses then do not free this memory but
-	 * mark them not present - any buggy init-section access will
-	 * create a kernel page fault:
-	 */
-#ifdef CONFIG_DEBUG_PAGEALLOC
-	printk(KERN_INFO "debug: unmapping init memory %08lx..%08lx\n",
-		begin, PAGE_ALIGN(end));
-	set_memory_np(begin, (end - begin) >> PAGE_SHIFT);
-#else
-	printk(KERN_INFO "Freeing %s: %luk freed\n", what, (end - begin) >> 10);
-
-	for (; addr < end; addr += PAGE_SIZE) {
-		ClearPageReserved(virt_to_page(addr));
-		init_page_count(virt_to_page(addr));
-		memset((void *)(addr & ~(PAGE_SIZE-1)),
-			POISON_FREE_INITMEM, PAGE_SIZE);
-		free_page(addr);
-		totalram_pages++;
-	}
-#endif
-}
-
-void free_initmem(void)
-{
-	free_init_pages("unused kernel memory",
-			(unsigned long)(&__init_begin),
-			(unsigned long)(&__init_end));
 }
 
 #ifdef CONFIG_DEBUG_RODATA
@@ -818,6 +950,13 @@ EXPORT_SYMBOL_GPL(rodata_test_data);
 void mark_rodata_ro(void)
 {
 	unsigned long start = PFN_ALIGN(_stext), end = PFN_ALIGN(__end_rodata);
+	unsigned long rodata_start =
+		((unsigned long)__start_rodata + PAGE_SIZE - 1) & PAGE_MASK;
+
+#ifdef CONFIG_DYNAMIC_FTRACE
+	/* Dynamic tracing modifies the kernel text section */
+	start = rodata_start;
+#endif
 
 	printk(KERN_INFO "Write protecting the kernel read-only data: %luk\n",
 	       (end - start) >> 10);
@@ -827,8 +966,7 @@ void mark_rodata_ro(void)
 	 * The rodata section (but not the kernel text!) should also be
 	 * not-executable.
 	 */
-	start = ((unsigned long)__start_rodata + PAGE_SIZE - 1) & PAGE_MASK;
-	set_memory_nx(start, (end - start) >> PAGE_SHIFT);
+	set_memory_nx(rodata_start, (end - rodata_start) >> PAGE_SHIFT);
 
 	rodata_test();
 
@@ -843,13 +981,6 @@ void mark_rodata_ro(void)
 
 #endif
 
-#ifdef CONFIG_BLK_DEV_INITRD
-void free_initrd_mem(unsigned long start, unsigned long end)
-{
-	free_init_pages("initrd memory", start, end);
-}
-#endif
-
 int __init reserve_bootmem_generic(unsigned long phys, unsigned long len,
 				   int flags)
 {
@@ -1037,9 +1168,6 @@ vmemmap_populate(struct page *start_page, unsigned long size, int node)
 						PAGE_KERNEL_LARGE);
 				set_pmd(pmd, __pmd(pte_val(entry)));
 
-				addr_end = addr + PMD_SIZE;
-				p_end = p + PMD_SIZE;
-
 				/* check to see if we have contiguous blocks */
 				if (p_end != p || node_start != node) {
 					if (p_start)
@@ -1049,6 +1177,9 @@ vmemmap_populate(struct page *start_page, unsigned long size, int node)
 					node_start = node;
 					p_start = p;
 				}
+
+				addr_end = addr + PMD_SIZE;
+				p_end = p + PMD_SIZE;
 			} else
 				vmemmap_verify((pte_t *)pmd, node, addr, next);
 		}