Merge branch 'x86/mm' into x86/core

1 files changed, 59 insertions, 196 deletions

diff --git a/arch/x86/mm/init_32.c b/arch/x86/mm/init_32.c
index 47df0e1bbeb9..2966c6b8d304 100644
--- a/arch/x86/mm/init_32.c
+++ b/arch/x86/mm/init_32.c
@@ -49,6 +49,7 @@
 #include <asm/paravirt.h>
 #include <asm/setup.h>
 #include <asm/cacheflush.h>
+#include <asm/init.h>
 
 unsigned long max_low_pfn_mapped;
 unsigned long max_pfn_mapped;
@@ -58,19 +59,14 @@ unsigned long highstart_pfn, highend_pfn;
 
 static noinline int do_test_wp_bit(void);
 
-
-static unsigned long __initdata table_start;
-static unsigned long __meminitdata table_end;
-static unsigned long __meminitdata table_top;
-
-static int __initdata after_init_bootmem;
+bool __read_mostly __vmalloc_start_set = false;
 
 static __init void *alloc_low_page(void)
 {
-	unsigned long pfn = table_end++;
+	unsigned long pfn = e820_table_end++;
 	void *adr;
 
-	if (pfn >= table_top)
+	if (pfn >= e820_table_top)
 		panic("alloc_low_page: ran out of memory");
 
 	adr = __va(pfn * PAGE_SIZE);
@@ -90,7 +86,7 @@ static pmd_t * __init one_md_table_init(pgd_t *pgd)
 
 #ifdef CONFIG_X86_PAE
 	if (!(pgd_val(*pgd) & _PAGE_PRESENT)) {
-		if (after_init_bootmem)
+		if (after_bootmem)
 			pmd_table = (pmd_t *)alloc_bootmem_low_pages(PAGE_SIZE);
 		else
 			pmd_table = (pmd_t *)alloc_low_page();
@@ -117,7 +113,7 @@ static pte_t * __init one_page_table_init(pmd_t *pmd)
 	if (!(pmd_val(*pmd) & _PAGE_PRESENT)) {
 		pte_t *page_table = NULL;
 
-		if (after_init_bootmem) {
+		if (after_bootmem) {
 #ifdef CONFIG_DEBUG_PAGEALLOC
 			page_table = (pte_t *) alloc_bootmem_pages(PAGE_SIZE);
 #endif
@@ -168,12 +164,12 @@ static pte_t *__init page_table_kmap_check(pte_t *pte, pmd_t *pmd,
 	if (pmd_idx_kmap_begin != pmd_idx_kmap_end
 	    && (vaddr >> PMD_SHIFT) >= pmd_idx_kmap_begin
 	    && (vaddr >> PMD_SHIFT) <= pmd_idx_kmap_end
-	    && ((__pa(pte) >> PAGE_SHIFT) < table_start
-		|| (__pa(pte) >> PAGE_SHIFT) >= table_end)) {
+	    && ((__pa(pte) >> PAGE_SHIFT) < e820_table_start
+		|| (__pa(pte) >> PAGE_SHIFT) >= e820_table_end)) {
 		pte_t *newpte;
 		int i;
 
-		BUG_ON(after_init_bootmem);
+		BUG_ON(after_bootmem);
 		newpte = alloc_low_page();
 		for (i = 0; i < PTRS_PER_PTE; i++)
 			set_pte(newpte + i, pte[i]);
@@ -242,11 +238,14 @@ static inline int is_kernel_text(unsigned long addr)
  * of max_low_pfn pages, by creating page tables starting from address
  * PAGE_OFFSET:
  */
-static void __init kernel_physical_mapping_init(pgd_t *pgd_base,
-						unsigned long start_pfn,
-						unsigned long end_pfn,
-						int use_pse)
+unsigned long __init
+kernel_physical_mapping_init(unsigned long start,
+			     unsigned long end,
+			     unsigned long page_size_mask)
 {
+	int use_pse = page_size_mask == (1<<PG_LEVEL_2M);
+	unsigned long start_pfn, end_pfn;
+	pgd_t *pgd_base = swapper_pg_dir;
 	int pgd_idx, pmd_idx, pte_ofs;
 	unsigned long pfn;
 	pgd_t *pgd;
@@ -255,6 +254,9 @@ static void __init kernel_physical_mapping_init(pgd_t *pgd_base,
 	unsigned pages_2m, pages_4k;
 	int mapping_iter;
 
+	start_pfn = start >> PAGE_SHIFT;
+	end_pfn = end >> PAGE_SHIFT;
+
 	/*
 	 * First iteration will setup identity mapping using large/small pages
 	 * based on use_pse, with other attributes same as set by
@@ -369,26 +371,6 @@ repeat:
 		mapping_iter = 2;
 		goto repeat;
 	}
-}
-
-/*
- * 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 (iomem_is_exclusive(pagenr << PAGE_SHIFT))
-		return 0;
-	if (!page_is_ram(pagenr))
-		return 1;
 	return 0;
 }
 
@@ -545,8 +527,9 @@ void __init native_pagetable_setup_done(pgd_t *base)
  * be partially populated, and so it avoids stomping on any existing
  * mappings.
  */
-static void __init early_ioremap_page_table_range_init(pgd_t *pgd_base)
+void __init early_ioremap_page_table_range_init(void)
 {
+	pgd_t *pgd_base = swapper_pg_dir;
 	unsigned long vaddr, end;
 
 	/*
@@ -641,7 +624,7 @@ static int __init noexec_setup(char *str)
 }
 early_param("noexec", noexec_setup);
 
-static void __init set_nx(void)
+void __init set_nx(void)
 {
 	unsigned int v[4], l, h;
 
@@ -793,6 +776,8 @@ void __init initmem_init(unsigned long start_pfn,
 #ifdef CONFIG_FLATMEM
 	max_mapnr = num_physpages;
 #endif
+	__vmalloc_start_set = true;
+
 	printk(KERN_NOTICE "%ldMB LOWMEM available.\n",
 			pages_to_mb(max_low_pfn));
 
@@ -814,176 +799,61 @@ static void __init zone_sizes_init(void)
 	free_area_init_nodes(max_zone_pfns);
 }
 
+static unsigned long __init setup_node_bootmem(int nodeid,
+				 unsigned long start_pfn,
+				 unsigned long end_pfn,
+				 unsigned long bootmap)
+{
+	unsigned long bootmap_size;
+
+	if (start_pfn > max_low_pfn)
+		return bootmap;
+	if (end_pfn > max_low_pfn)
+		end_pfn = max_low_pfn;
+
+	/* don't touch min_low_pfn */
+	bootmap_size = init_bootmem_node(NODE_DATA(nodeid),
+					 bootmap >> PAGE_SHIFT,
+					 start_pfn, end_pfn);
+	printk(KERN_INFO "  node %d low ram: %08lx - %08lx\n",
+		nodeid, start_pfn<<PAGE_SHIFT, end_pfn<<PAGE_SHIFT);
+	printk(KERN_INFO "  node %d bootmap %08lx - %08lx\n",
+		 nodeid, bootmap, bootmap + bootmap_size);
+	free_bootmem_with_active_regions(nodeid, end_pfn);
+	early_res_to_bootmem(start_pfn<<PAGE_SHIFT, end_pfn<<PAGE_SHIFT);
+
+	return bootmap + bootmap_size;
+}
+
 void __init setup_bootmem_allocator(void)
 {
-	int i;
+	int nodeid;
 	unsigned long bootmap_size, bootmap;
 	/*
 	 * Initialize the boot-time allocator (with low memory only):
 	 */
 	bootmap_size = bootmem_bootmap_pages(max_low_pfn)<<PAGE_SHIFT;
-	bootmap = find_e820_area(min_low_pfn<<PAGE_SHIFT,
-				 max_pfn_mapped<<PAGE_SHIFT, bootmap_size,
+	bootmap = find_e820_area(0, max_pfn_mapped<<PAGE_SHIFT, bootmap_size,
 				 PAGE_SIZE);
 	if (bootmap == -1L)
 		panic("Cannot find bootmem map of size %ld\n", bootmap_size);
 	reserve_early(bootmap, bootmap + bootmap_size, "BOOTMAP");
 
-	/* don't touch min_low_pfn */
-	bootmap_size = init_bootmem_node(NODE_DATA(0), bootmap >> PAGE_SHIFT,
-					 min_low_pfn, max_low_pfn);
 	printk(KERN_INFO "  mapped low ram: 0 - %08lx\n",
 		 max_pfn_mapped<<PAGE_SHIFT);
-	printk(KERN_INFO "  low ram: %08lx - %08lx\n",
-		 min_low_pfn<<PAGE_SHIFT, max_low_pfn<<PAGE_SHIFT);
-	printk(KERN_INFO "  bootmap %08lx - %08lx\n",
-		 bootmap, bootmap + bootmap_size);
-	for_each_online_node(i)
-		free_bootmem_with_active_regions(i, max_low_pfn);
-	early_res_to_bootmem(0, max_low_pfn<<PAGE_SHIFT);
-
-	after_init_bootmem = 1;
-}
-
-static void __init find_early_table_space(unsigned long end, int use_pse)
-{
-	unsigned long puds, pmds, ptes, tables, start;
-
-	puds = (end + PUD_SIZE - 1) >> PUD_SHIFT;
-	tables = roundup(puds * sizeof(pud_t), PAGE_SIZE);
-
-	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);
-		extra += PMD_SIZE;
-		ptes = (extra + PAGE_SIZE - 1) >> PAGE_SHIFT;
-	} else
-		ptes = (end + PAGE_SIZE - 1) >> PAGE_SHIFT;
+	printk(KERN_INFO "  low ram: 0 - %08lx\n", max_low_pfn<<PAGE_SHIFT);
 
-	tables += roundup(ptes * sizeof(pte_t), PAGE_SIZE);
-
-	/* for fixmap */
-	tables += roundup(__end_of_fixed_addresses * sizeof(pte_t), PAGE_SIZE);
-
-	/*
-	 * 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.
-	 */
-	start = 0x7000;
-	table_start = find_e820_area(start, max_pfn_mapped<<PAGE_SHIFT,
-					tables, PAGE_SIZE);
-	if (table_start == -1UL)
-		panic("Cannot find space for the kernel page tables");
-
-	table_start >>= PAGE_SHIFT;
-	table_end = table_start;
-	table_top = table_start + (tables>>PAGE_SHIFT);
-
-	printk(KERN_DEBUG "kernel direct mapping tables up to %lx @ %lx-%lx\n",
-		end, table_start << PAGE_SHIFT,
-		(table_start << PAGE_SHIFT) + tables);
-}
-
-unsigned long __init_refok init_memory_mapping(unsigned long start,
-						unsigned long end)
-{
-	pgd_t *pgd_base = swapper_pg_dir;
-	unsigned long start_pfn, end_pfn;
-	unsigned long big_page_start;
-#ifdef CONFIG_DEBUG_PAGEALLOC
-	/*
-	 * 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.
-	 */
-	int use_pse = 0;
+#ifdef CONFIG_NEED_MULTIPLE_NODES
+	for_each_online_node(nodeid)
+		bootmap = setup_node_bootmem(nodeid, node_start_pfn[nodeid],
+					node_end_pfn[nodeid], bootmap);
 #else
-	int use_pse = cpu_has_pse;
-#endif
-
-	/*
-	 * Find space for the kernel direct mapping tables.
-	 */
-	if (!after_init_bootmem)
-		find_early_table_space(end, use_pse);
-
-#ifdef CONFIG_X86_PAE
-	set_nx();
-	if (nx_enabled)
-		printk(KERN_INFO "NX (Execute Disable) protection: active\n");
+	bootmap = setup_node_bootmem(0, 0, max_low_pfn, bootmap);
 #endif
 
-	/* Enable PSE if available */
-	if (cpu_has_pse)
-		set_in_cr4(X86_CR4_PSE);
-
-	/* Enable PGE if available */
-	if (cpu_has_pge) {
-		set_in_cr4(X86_CR4_PGE);
-		__supported_pte_mask |= _PAGE_GLOBAL;
-	}
-
-	/*
-	 * Don't use a large page for the first 2/4MB of memory
-	 * because there are often fixed size MTRRs in there
-	 * and overlapping MTRRs into large pages can cause
-	 * slowdowns.
-	 */
-	big_page_start = PMD_SIZE;
-
-	if (start < big_page_start) {
-		start_pfn = start >> PAGE_SHIFT;
-		end_pfn = min(big_page_start>>PAGE_SHIFT, end>>PAGE_SHIFT);
-	} else {
-		/* head is not big page alignment ? */
-		start_pfn = start >> PAGE_SHIFT;
-		end_pfn = ((start + (PMD_SIZE - 1))>>PMD_SHIFT)
-				 << (PMD_SHIFT - PAGE_SHIFT);
-	}
-	if (start_pfn < end_pfn)
-		kernel_physical_mapping_init(pgd_base, start_pfn, end_pfn, 0);
-
-	/* big page range */
-	start_pfn = ((start + (PMD_SIZE - 1))>>PMD_SHIFT)
-			 << (PMD_SHIFT - PAGE_SHIFT);
-	if (start_pfn < (big_page_start >> PAGE_SHIFT))
-		start_pfn =  big_page_start >> PAGE_SHIFT;
-	end_pfn = (end>>PMD_SHIFT) << (PMD_SHIFT - PAGE_SHIFT);
-	if (start_pfn < end_pfn)
-		kernel_physical_mapping_init(pgd_base, start_pfn, end_pfn,
-					     use_pse);
-
-	/* tail is not big page alignment ? */
-	start_pfn = end_pfn;
-	if (start_pfn > (big_page_start>>PAGE_SHIFT)) {
-		end_pfn = end >> PAGE_SHIFT;
-		if (start_pfn < end_pfn)
-			kernel_physical_mapping_init(pgd_base, start_pfn,
-							 end_pfn, 0);
-	}
-
-	early_ioremap_page_table_range_init(pgd_base);
-
-	load_cr3(swapper_pg_dir);
-
-	__flush_tlb_all();
-
-	if (!after_init_bootmem)
-		reserve_early(table_start << PAGE_SHIFT,
-				 table_end << PAGE_SHIFT, "PGTABLE");
-
-	if (!after_init_bootmem)
-		early_memtest(start, end);
-
-	return end >> PAGE_SHIFT;
+	after_bootmem = 1;
 }
 
-
 /*
  * paging_init() sets up the page tables - note that the first 8MB are
  * already mapped by head.S.
@@ -1217,13 +1087,6 @@ 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)
 {