Merge branch 'x86-boot-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip

Pull x86 boot updates from Ingo Molnar:
 "The changes in this cycle were:

   - Save e820 table RAM footprint on larger kernel configurations.
     (Denys Vlasenko)

   - pmem related fixes (Dan Williams)

   - theoretical e820 boundary condition fix (Wei Yang)"

* 'x86-boot-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
  x86/boot: Fix kdump, cleanup aborted E820_PRAM max_pfn manipulation
  x86/e820: Use much less memory for e820/e820_saved, save up to 120k
  x86/e820: Prepare e280 code for switch to dynamic storage
  x86/e820: Mark some static functions __init
  x86/e820: Fix very large 'size' handling boundary condition

arch/x86/include/asm/e820.h

@@ -10,8 +10,8 @@
 #include <uapi/asm/e820.h>
 #ifndef __ASSEMBLY__
 /* see comment in arch/x86/kernel/e820.c */
-extern struct e820map e820;
-extern struct e820map e820_saved;
+extern struct e820map *e820;
+extern struct e820map *e820_saved;
 
 extern unsigned long pci_mem_start;
 extern int e820_any_mapped(u64 start, u64 end, unsigned type);
@@ -53,6 +53,8 @@ extern void e820_reserve_resources_late(void);
 extern void setup_memory_map(void);
 extern char *default_machine_specific_memory_setup(void);
 
+extern void e820_reallocate_tables(void);
+
 /*
  * Returns true iff the specified range [s,e) is completely contained inside
  * the ISA region.

arch/x86/kernel/e820.c

@@ -40,8 +40,10 @@
  * user can e.g. boot the original kernel with mem=1G while still booting the
  * next kernel with full memory.
  */
-struct e820map e820;
-struct e820map e820_saved;
+static struct e820map initial_e820  __initdata;
+static struct e820map initial_e820_saved  __initdata;
+struct e820map *e820 __refdata = &initial_e820;
+struct e820map *e820_saved __refdata = &initial_e820_saved;
 
 /* For PCI or other memory-mapped resources */
 unsigned long pci_mem_start = 0xaeedbabe;
@@ -58,8 +60,8 @@ e820_any_mapped(u64 start, u64 end, unsigned type)
 {
 	int i;
 
-	for (i = 0; i < e820.nr_map; i++) {
-		struct e820entry *ei = &e820.map[i];
+	for (i = 0; i < e820->nr_map; i++) {
+		struct e820entry *ei = &e820->map[i];
 
 		if (type && ei->type != type)
 			continue;
@@ -81,8 +83,8 @@ int __init e820_all_mapped(u64 start, u64 end, unsigned type)
 {
 	int i;
 
-	for (i = 0; i < e820.nr_map; i++) {
-		struct e820entry *ei = &e820.map[i];
+	for (i = 0; i < e820->nr_map; i++) {
+		struct e820entry *ei = &e820->map[i];
 
 		if (type && ei->type != type)
 			continue;
@@ -128,7 +130,7 @@ static void __init __e820_add_region(struct e820map *e820x, u64 start, u64 size,
 
 void __init e820_add_region(u64 start, u64 size, int type)
 {
-	__e820_add_region(&e820, start, size, type);
+	__e820_add_region(e820, start, size, type);
 }
 
 static void __init e820_print_type(u32 type)
@@ -164,12 +166,12 @@ void __init e820_print_map(char *who)
 {
 	int i;
 
-	for (i = 0; i < e820.nr_map; i++) {
+	for (i = 0; i < e820->nr_map; i++) {
 		printk(KERN_INFO "%s: [mem %#018Lx-%#018Lx] ", who,
-		       (unsigned long long) e820.map[i].addr,
+		       (unsigned long long) e820->map[i].addr,
 		       (unsigned long long)
-		       (e820.map[i].addr + e820.map[i].size - 1));
-		e820_print_type(e820.map[i].type);
+		       (e820->map[i].addr + e820->map[i].size - 1));
+		e820_print_type(e820->map[i].type);
 		printk(KERN_CONT "\n");
 	}
 }
@@ -348,7 +350,7 @@ int __init sanitize_e820_map(struct e820entry *biosmap, int max_nr_map,
 		 * continue building up new bios map based on this
 		 * information
 		 */
-		if (current_type != last_type || current_type == E820_PRAM) {
+		if (current_type != last_type) {
 			if (last_type != 0)	 {
 				new_bios[new_bios_entry].size =
 					change_point[chgidx]->addr - last_addr;
@@ -388,11 +390,11 @@ static int __init __append_e820_map(struct e820entry *biosmap, int nr_map)
 	while (nr_map) {
 		u64 start = biosmap->addr;
 		u64 size = biosmap->size;
-		u64 end = start + size;
+		u64 end = start + size - 1;
 		u32 type = biosmap->type;
 
 		/* Overflow in 64 bits? Ignore the memory map. */
-		if (start > end)
+		if (start > end && likely(size))
 			return -1;
 
 		e820_add_region(start, size, type);
@@ -493,13 +495,13 @@ static u64 __init __e820_update_range(struct e820map *e820x, u64 start,
 u64 __init e820_update_range(u64 start, u64 size, unsigned old_type,
 			     unsigned new_type)
 {
-	return __e820_update_range(&e820, start, size, old_type, new_type);
+	return __e820_update_range(e820, start, size, old_type, new_type);
 }
 
 static u64 __init e820_update_range_saved(u64 start, u64 size,
 					  unsigned old_type, unsigned new_type)
 {
-	return __e820_update_range(&e820_saved, start, size, old_type,
+	return __e820_update_range(e820_saved, start, size, old_type,
 				     new_type);
 }
 
@@ -521,8 +523,8 @@ u64 __init e820_remove_range(u64 start, u64 size, unsigned old_type,
 		e820_print_type(old_type);
 	printk(KERN_CONT "\n");
 
-	for (i = 0; i < e820.nr_map; i++) {
-		struct e820entry *ei = &e820.map[i];
+	for (i = 0; i < e820->nr_map; i++) {
+		struct e820entry *ei = &e820->map[i];
 		u64 final_start, final_end;
 		u64 ei_end;
 
@@ -566,15 +568,15 @@ u64 __init e820_remove_range(u64 start, u64 size, unsigned old_type,
 
 void __init update_e820(void)
 {
-	if (sanitize_e820_map(e820.map, ARRAY_SIZE(e820.map), &e820.nr_map))
+	if (sanitize_e820_map(e820->map, ARRAY_SIZE(e820->map), &e820->nr_map))
 		return;
 	printk(KERN_INFO "e820: modified physical RAM map:\n");
 	e820_print_map("modified");
 }
 static void __init update_e820_saved(void)
 {
-	sanitize_e820_map(e820_saved.map, ARRAY_SIZE(e820_saved.map),
-				&e820_saved.nr_map);
+	sanitize_e820_map(e820_saved->map, ARRAY_SIZE(e820_saved->map),
+				&e820_saved->nr_map);
 }
 #define MAX_GAP_END 0x100000000ull
 /*
@@ -584,14 +586,14 @@ __init int e820_search_gap(unsigned long *gapstart, unsigned long *gapsize,
 		unsigned long start_addr, unsigned long long end_addr)
 {
 	unsigned long long last;
-	int i = e820.nr_map;
+	int i = e820->nr_map;
 	int found = 0;
 
 	last = (end_addr && end_addr < MAX_GAP_END) ? end_addr : MAX_GAP_END;
 
 	while (--i >= 0) {
-		unsigned long long start = e820.map[i].addr;
-		unsigned long long end = start + e820.map[i].size;
+		unsigned long long start = e820->map[i].addr;
+		unsigned long long end = start + e820->map[i].size;
 
 		if (end < start_addr)
 			continue;
@@ -649,6 +651,33 @@ __init void e820_setup_gap(void)
 	       gapstart, gapstart + gapsize - 1);
 }
 
+/*
+ * Called late during init, in free_initmem().
+ *
+ * Initial e820 and e820_saved are largish __initdata arrays.
+ * Copy them to (usually much smaller) dynamically allocated area.
+ * This is done after all tweaks we ever do to them:
+ * all functions which modify them are __init functions,
+ * they won't exist after this point.
+ */
+__init void e820_reallocate_tables(void)
+{
+	struct e820map *n;
+	int size;
+
+	size = offsetof(struct e820map, map) + sizeof(struct e820entry) * e820->nr_map;
+	n = kmalloc(size, GFP_KERNEL);
+	BUG_ON(!n);
+	memcpy(n, e820, size);
+	e820 = n;
+
+	size = offsetof(struct e820map, map) + sizeof(struct e820entry) * e820_saved->nr_map;
+	n = kmalloc(size, GFP_KERNEL);
+	BUG_ON(!n);
+	memcpy(n, e820_saved, size);
+	e820_saved = n;
+}
+
 /**
  * Because of the size limitation of struct boot_params, only first
  * 128 E820 memory entries are passed to kernel via
@@ -665,7 +694,7 @@ void __init parse_e820_ext(u64 phys_addr, u32 data_len)
 	entries = sdata->len / sizeof(struct e820entry);
 	extmap = (struct e820entry *)(sdata->data);
 	__append_e820_map(extmap, entries);
-	sanitize_e820_map(e820.map, ARRAY_SIZE(e820.map), &e820.nr_map);
+	sanitize_e820_map(e820->map, ARRAY_SIZE(e820->map), &e820->nr_map);
 	early_memunmap(sdata, data_len);
 	printk(KERN_INFO "e820: extended physical RAM map:\n");
 	e820_print_map("extended");
@@ -686,8 +715,8 @@ void __init e820_mark_nosave_regions(unsigned long limit_pfn)
 	int i;
 	unsigned long pfn = 0;
 
-	for (i = 0; i < e820.nr_map; i++) {
-		struct e820entry *ei = &e820.map[i];
+	for (i = 0; i < e820->nr_map; i++) {
+		struct e820entry *ei = &e820->map[i];
 
 		if (pfn < PFN_UP(ei->addr))
 			register_nosave_region(pfn, PFN_UP(ei->addr));
@@ -712,8 +741,8 @@ static int __init e820_mark_nvs_memory(void)
 {
 	int i;
 
-	for (i = 0; i < e820.nr_map; i++) {
-		struct e820entry *ei = &e820.map[i];
+	for (i = 0; i < e820->nr_map; i++) {
+		struct e820entry *ei = &e820->map[i];
 
 		if (ei->type == E820_NVS)
 			acpi_nvs_register(ei->addr, ei->size);
@@ -754,22 +783,18 @@ u64 __init early_reserve_e820(u64 size, u64 align)
 /*
  * Find the highest page frame number we have available
  */
-static unsigned long __init e820_end_pfn(unsigned long limit_pfn)
+static unsigned long __init e820_end_pfn(unsigned long limit_pfn, unsigned type)
 {
 	int i;
 	unsigned long last_pfn = 0;
 	unsigned long max_arch_pfn = MAX_ARCH_PFN;
 
-	for (i = 0; i < e820.nr_map; i++) {
-		struct e820entry *ei = &e820.map[i];
+	for (i = 0; i < e820->nr_map; i++) {
+		struct e820entry *ei = &e820->map[i];
 		unsigned long start_pfn;
 		unsigned long end_pfn;
 
-		/*
-		 * Persistent memory is accounted as ram for purposes of
-		 * establishing max_pfn and mem_map.
-		 */
-		if (ei->type != E820_RAM && ei->type != E820_PRAM)
+		if (ei->type != type)
 			continue;
 
 		start_pfn = ei->addr >> PAGE_SHIFT;
@@ -794,15 +819,15 @@ static unsigned long __init e820_end_pfn(unsigned long limit_pfn)
 }
 unsigned long __init e820_end_of_ram_pfn(void)
 {
-	return e820_end_pfn(MAX_ARCH_PFN);
+	return e820_end_pfn(MAX_ARCH_PFN, E820_RAM);
 }
 
 unsigned long __init e820_end_of_low_ram_pfn(void)
 {
-	return e820_end_pfn(1UL << (32-PAGE_SHIFT));
+	return e820_end_pfn(1UL << (32 - PAGE_SHIFT), E820_RAM);
 }
 
-static void early_panic(char *msg)
+static void __init early_panic(char *msg)
 {
 	early_printk(msg);
 	panic(msg);
@@ -856,7 +881,7 @@ static int __init parse_memmap_one(char *p)
 		 */
 		saved_max_pfn = e820_end_of_ram_pfn();
 #endif
-		e820.nr_map = 0;
+		e820->nr_map = 0;
 		userdef = 1;
 		return 0;
 	}
@@ -903,8 +928,8 @@ early_param("memmap", parse_memmap_opt);
 void __init finish_e820_parsing(void)
 {
 	if (userdef) {
-		if (sanitize_e820_map(e820.map, ARRAY_SIZE(e820.map),
-					&e820.nr_map) < 0)
+		if (sanitize_e820_map(e820->map, ARRAY_SIZE(e820->map),
+					&e820->nr_map) < 0)
 			early_panic("Invalid user supplied memory map");
 
 		printk(KERN_INFO "e820: user-defined physical RAM map:\n");
@@ -912,7 +937,7 @@ void __init finish_e820_parsing(void)
 	}
 }
 
-static const char *e820_type_to_string(int e820_type)
+static const char *__init e820_type_to_string(int e820_type)
 {
 	switch (e820_type) {
 	case E820_RESERVED_KERN:
@@ -926,7 +951,7 @@ static const char *e820_type_to_string(int e820_type)
 	}
 }
 
-static unsigned long e820_type_to_iomem_type(int e820_type)
+static unsigned long __init e820_type_to_iomem_type(int e820_type)
 {
 	switch (e820_type) {
 	case E820_RESERVED_KERN:
@@ -942,7 +967,7 @@ static unsigned long e820_type_to_iomem_type(int e820_type)
 	}
 }
 
-static unsigned long e820_type_to_iores_desc(int e820_type)
+static unsigned long __init e820_type_to_iores_desc(int e820_type)
 {
 	switch (e820_type) {
 	case E820_ACPI:
@@ -961,7 +986,7 @@ static unsigned long e820_type_to_iores_desc(int e820_type)
 	}
 }
 
-static bool do_mark_busy(u32 type, struct resource *res)
+static bool __init do_mark_busy(u32 type, struct resource *res)
 {
 	/* this is the legacy bios/dos rom-shadow + mmio region */
 	if (res->start < (1ULL<<20))
@@ -991,35 +1016,35 @@ void __init e820_reserve_resources(void)
 	struct resource *res;
 	u64 end;
 
-	res = alloc_bootmem(sizeof(struct resource) * e820.nr_map);
+	res = alloc_bootmem(sizeof(struct resource) * e820->nr_map);
 	e820_res = res;
-	for (i = 0; i < e820.nr_map; i++) {
-		end = e820.map[i].addr + e820.map[i].size - 1;
+	for (i = 0; i < e820->nr_map; i++) {
+		end = e820->map[i].addr + e820->map[i].size - 1;
 		if (end != (resource_size_t)end) {
 			res++;
 			continue;
 		}
-		res->name = e820_type_to_string(e820.map[i].type);
-		res->start = e820.map[i].addr;
+		res->name = e820_type_to_string(e820->map[i].type);
+		res->start = e820->map[i].addr;
 		res->end = end;
 
-		res->flags = e820_type_to_iomem_type(e820.map[i].type);
-		res->desc = e820_type_to_iores_desc(e820.map[i].type);
+		res->flags = e820_type_to_iomem_type(e820->map[i].type);
+		res->desc = e820_type_to_iores_desc(e820->map[i].type);
 
 		/*
 		 * don't register the region that could be conflicted with
 		 * pci device BAR resource and insert them later in
 		 * pcibios_resource_survey()
 		 */
-		if (do_mark_busy(e820.map[i].type, res)) {
+		if (do_mark_busy(e820->map[i].type, res)) {
 			res->flags |= IORESOURCE_BUSY;
 			insert_resource(&iomem_resource, res);
 		}
 		res++;
 	}
 
-	for (i = 0; i < e820_saved.nr_map; i++) {
-		struct e820entry *entry = &e820_saved.map[i];
+	for (i = 0; i < e820_saved->nr_map; i++) {
+		struct e820entry *entry = &e820_saved->map[i];
 		firmware_map_add_early(entry->addr,
 			entry->addr + entry->size,
 			e820_type_to_string(entry->type));
@@ -1027,7 +1052,7 @@ void __init e820_reserve_resources(void)
 }
 
 /* How much should we pad RAM ending depending on where it is? */
-static unsigned long ram_alignment(resource_size_t pos)
+static unsigned long __init ram_alignment(resource_size_t pos)
 {
 	unsigned long mb = pos >> 20;
 
@@ -1051,7 +1076,7 @@ void __init e820_reserve_resources_late(void)
 	struct resource *res;
 
 	res = e820_res;
-	for (i = 0; i < e820.nr_map; i++) {
+	for (i = 0; i < e820->nr_map; i++) {
 		if (!res->parent && res->end)
 			insert_resource_expand_to_fit(&iomem_resource, res);
 		res++;
@@ -1061,8 +1086,8 @@ void __init e820_reserve_resources_late(void)
 	 * Try to bump up RAM regions to reasonable boundaries to
 	 * avoid stolen RAM:
 	 */
-	for (i = 0; i < e820.nr_map; i++) {
-		struct e820entry *entry = &e820.map[i];
+	for (i = 0; i < e820->nr_map; i++) {
+		struct e820entry *entry = &e820->map[i];
 		u64 start, end;
 
 		if (entry->type != E820_RAM)
@@ -1110,7 +1135,7 @@ char *__init default_machine_specific_memory_setup(void)
 			who = "BIOS-e801";
 		}
 
-		e820.nr_map = 0;
+		e820->nr_map = 0;
 		e820_add_region(0, LOWMEMSIZE(), E820_RAM);
 		e820_add_region(HIGH_MEMORY, mem_size << 10, E820_RAM);
 	}
@@ -1124,7 +1149,7 @@ void __init setup_memory_map(void)
 	char *who;
 
 	who = x86_init.resources.memory_setup();
-	memcpy(&e820_saved, &e820, sizeof(struct e820map));
+	memcpy(e820_saved, e820, sizeof(struct e820map));
 	printk(KERN_INFO "e820: BIOS-provided physical RAM map:\n");
 	e820_print_map(who);
 }
@@ -1141,8 +1166,8 @@ void __init memblock_x86_fill(void)
 	 */
 	memblock_allow_resize();
 
-	for (i = 0; i < e820.nr_map; i++) {
-		struct e820entry *ei = &e820.map[i];
+	for (i = 0; i < e820->nr_map; i++) {
+		struct e820entry *ei = &e820->map[i];
 
 		end = ei->addr + ei->size;
 		if (end != (resource_size_t)end)

arch/x86/kernel/early-quirks.c

@@ -555,7 +555,7 @@ intel_graphics_stolen(int num, int slot, int func,
 
 	/* Mark this space as reserved */
 	e820_add_region(base, size, E820_RESERVED);
-	sanitize_e820_map(e820.map, ARRAY_SIZE(e820.map), &e820.nr_map);
+	sanitize_e820_map(e820->map, ARRAY_SIZE(e820->map), &e820->nr_map);
 }
 
 static void __init intel_graphics_quirks(int num, int slot, int func)

arch/x86/kernel/kexec-bzimage64.c

@@ -99,14 +99,14 @@ static int setup_e820_entries(struct boot_params *params)
 {
 	unsigned int nr_e820_entries;
 
-	nr_e820_entries = e820_saved.nr_map;
+	nr_e820_entries = e820_saved->nr_map;
 
 	/* TODO: Pass entries more than E820MAX in bootparams setup data */
 	if (nr_e820_entries > E820MAX)
 		nr_e820_entries = E820MAX;
 
 	params->e820_entries = nr_e820_entries;
-	memcpy(&params->e820_map, &e820_saved.map,
+	memcpy(&params->e820_map, &e820_saved->map,
 	       nr_e820_entries * sizeof(struct e820entry));
 
 	return 0;

arch/x86/kernel/resource.c

@@ -27,8 +27,8 @@ static void remove_e820_regions(struct resource *avail)
 	int i;
 	struct e820entry *entry;
 
-	for (i = 0; i < e820.nr_map; i++) {
-		entry = &e820.map[i];
+	for (i = 0; i < e820->nr_map; i++) {
+		entry = &e820->map[i];
 
 		resource_clip(avail, entry->addr,
 			      entry->addr + entry->size - 1);

arch/x86/kernel/setup.c

@@ -458,8 +458,8 @@ static void __init e820_reserve_setup_data(void)
 		early_memunmap(data, sizeof(*data));
 	}
 
-	sanitize_e820_map(e820.map, ARRAY_SIZE(e820.map), &e820.nr_map);
-	memcpy(&e820_saved, &e820, sizeof(struct e820map));
+	sanitize_e820_map(e820->map, ARRAY_SIZE(e820->map), &e820->nr_map);
+	memcpy(e820_saved, e820, sizeof(struct e820map));
 	printk(KERN_INFO "extended physical RAM map:\n");
 	e820_print_map("reserve setup_data");
 }
@@ -763,7 +763,7 @@ static void __init trim_bios_range(void)
 	 */
 	e820_remove_range(BIOS_BEGIN, BIOS_END - BIOS_BEGIN, E820_RAM, 1);
 
-	sanitize_e820_map(e820.map, ARRAY_SIZE(e820.map), &e820.nr_map);
+	sanitize_e820_map(e820->map, ARRAY_SIZE(e820->map), &e820->nr_map);
 }
 
 /* called before trim_bios_range() to spare extra sanitize */
@@ -1032,7 +1032,7 @@ void __init setup_arch(char **cmdline_p)
 	if (ppro_with_ram_bug()) {
 		e820_update_range(0x70000000ULL, 0x40000ULL, E820_RAM,
 				  E820_RESERVED);
-		sanitize_e820_map(e820.map, ARRAY_SIZE(e820.map), &e820.nr_map);
+		sanitize_e820_map(e820->map, ARRAY_SIZE(e820->map), &e820->nr_map);
 		printk(KERN_INFO "fixed physical RAM map:\n");
 		e820_print_map("bad_ppro");
 	}

arch/x86/kernel/tboot.c

@@ -188,12 +188,12 @@ static int tboot_setup_sleep(void)
 
 	tboot->num_mac_regions = 0;
 
-	for (i = 0; i < e820.nr_map; i++) {
-		if ((e820.map[i].type != E820_RAM)
-		 && (e820.map[i].type != E820_RESERVED_KERN))
+	for (i = 0; i < e820->nr_map; i++) {
+		if ((e820->map[i].type != E820_RAM)
+		 && (e820->map[i].type != E820_RESERVED_KERN))
 			continue;
 
-		add_mac_region(e820.map[i].addr, e820.map[i].size);
+		add_mac_region(e820->map[i].addr, e820->map[i].size);
 	}
 
 	tboot->acpi_sinfo.kernel_s3_resume_vector =

arch/x86/mm/init.c

@@ -699,8 +699,10 @@ void free_init_pages(char *what, unsigned long begin, unsigned long end)
 	}
 }
 
-void free_initmem(void)
+void __ref free_initmem(void)
 {
+	e820_reallocate_tables();
+
 	free_init_pages("unused kernel",
 			(unsigned long)(&__init_begin),
 			(unsigned long)(&__init_end));

arch/x86/platform/efi/efi.c

@@ -166,7 +166,7 @@ static void __init do_add_efi_memmap(void)
 		}
 		e820_add_region(start, size, e820_type);
 	}
-	sanitize_e820_map(e820.map, ARRAY_SIZE(e820.map), &e820.nr_map);
+	sanitize_e820_map(e820->map, ARRAY_SIZE(e820->map), &e820->nr_map);
 }
 
 int __init efi_memblock_x86_reserve_range(void)

arch/x86/xen/setup.c

@@ -861,7 +861,7 @@ char * __init xen_memory_setup(void)
 	e820_add_region(ISA_START_ADDRESS, ISA_END_ADDRESS - ISA_START_ADDRESS,
 			E820_RESERVED);
 
-	sanitize_e820_map(e820.map, ARRAY_SIZE(e820.map), &e820.nr_map);
+	sanitize_e820_map(e820->map, ARRAY_SIZE(e820->map), &e820->nr_map);
 
 	/*
 	 * Check whether the kernel itself conflicts with the target E820 map.