setup.c 26.2 KB
Newer Older
L
Linus Torvalds 已提交
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25
/*
 *  Copyright (C) 1995  Linus Torvalds
 *
 *  Support of BIGMEM added by Gerhard Wichert, Siemens AG, July 1999
 *
 *  Memory region support
 *	David Parsons <orc@pell.chi.il.us>, July-August 1999
 *
 *  Added E820 sanitization routine (removes overlapping memory regions);
 *  Brian Moyle <bmoyle@mvista.com>, February 2001
 *
 * Moved CPU detection code to cpu/${cpu}.c
 *    Patrick Mochel <mochel@osdl.org>, March 2002
 *
 *  Provisions for empty E820 memory regions (reported by certain BIOSes).
 *  Alex Achenbach <xela@slit.de>, December 2002.
 *
 */

/*
 * This file handles the architecture-dependent parts of initialization
 */

#include <linux/sched.h>
#include <linux/mm.h>
26
#include <linux/mmzone.h>
27
#include <linux/screen_info.h>
L
Linus Torvalds 已提交
28 29
#include <linux/ioport.h>
#include <linux/acpi.h>
F
Feng Tang 已提交
30
#include <linux/sfi.h>
L
Linus Torvalds 已提交
31 32 33
#include <linux/apm_bios.h>
#include <linux/initrd.h>
#include <linux/bootmem.h>
34
#include <linux/memblock.h>
L
Linus Torvalds 已提交
35 36 37 38 39 40 41 42
#include <linux/seq_file.h>
#include <linux/console.h>
#include <linux/root_dev.h>
#include <linux/highmem.h>
#include <linux/module.h>
#include <linux/efi.h>
#include <linux/init.h>
#include <linux/edd.h>
43
#include <linux/iscsi_ibft.h>
L
Linus Torvalds 已提交
44
#include <linux/nodemask.h>
45
#include <linux/kexec.h>
46
#include <linux/dmi.h>
D
Dave Hansen 已提交
47
#include <linux/pfn.h>
48
#include <linux/pci.h>
49
#include <asm/pci-direct.h>
50
#include <linux/init_ohci1394_dma.h>
51
#include <linux/kvm_para.h>
52
#include <linux/dma-contiguous.h>
53

54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69
#include <linux/errno.h>
#include <linux/kernel.h>
#include <linux/stddef.h>
#include <linux/unistd.h>
#include <linux/ptrace.h>
#include <linux/user.h>
#include <linux/delay.h>

#include <linux/kallsyms.h>
#include <linux/cpufreq.h>
#include <linux/dma-mapping.h>
#include <linux/ctype.h>
#include <linux/uaccess.h>

#include <linux/percpu.h>
#include <linux/crash_dump.h>
70
#include <linux/tboot.h>
71
#include <linux/jiffies.h>
72

L
Linus Torvalds 已提交
73
#include <video/edid.h>
74

75
#include <asm/mtrr.h>
76
#include <asm/apic.h>
77
#include <asm/realmode.h>
L
Linus Torvalds 已提交
78 79 80
#include <asm/e820.h>
#include <asm/mpspec.h>
#include <asm/setup.h>
Y
Yinghai Lu 已提交
81
#include <asm/efi.h>
I
Ingo Molnar 已提交
82 83
#include <asm/timer.h>
#include <asm/i8259.h>
L
Linus Torvalds 已提交
84
#include <asm/sections.h>
85
#include <asm/dmi.h>
L
Linus Torvalds 已提交
86 87
#include <asm/io_apic.h>
#include <asm/ist.h>
88
#include <asm/setup_arch.h>
89
#include <asm/bios_ebda.h>
B
Bernhard Walle 已提交
90
#include <asm/cacheflush.h>
91
#include <asm/processor.h>
Y
Yinghai Lu 已提交
92
#include <asm/bugs.h>
L
Linus Torvalds 已提交
93

94
#include <asm/vsyscall.h>
95
#include <asm/cpu.h>
96 97
#include <asm/desc.h>
#include <asm/dma.h>
98
#include <asm/iommu.h>
99
#include <asm/gart.h>
100 101 102 103
#include <asm/mmu_context.h>
#include <asm/proto.h>

#include <asm/paravirt.h>
104
#include <asm/hypervisor.h>
105
#include <asm/olpc_ofw.h>
106 107 108 109

#include <asm/percpu.h>
#include <asm/topology.h>
#include <asm/apicdef.h>
110
#include <asm/amd_nb.h>
Y
Yinghai Lu 已提交
111 112 113
#ifdef CONFIG_X86_64
#include <asm/numa_64.h>
#endif
114
#include <asm/mce.h>
115
#include <asm/alternative.h>
116
#include <asm/prom.h>
117

118 119 120 121 122 123 124 125
/*
 * end_pfn only includes RAM, while max_pfn_mapped includes all e820 entries.
 * 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;

126
#ifdef CONFIG_DMI
127
RESERVE_BRK(dmi_alloc, 65536);
128
#endif
129

I
Ingo Molnar 已提交
130

131 132 133
static __initdata unsigned long _brk_start = (unsigned long)__brk_base;
unsigned long _brk_end = (unsigned long)__brk_base;

I
Ingo Molnar 已提交
134 135 136 137 138 139
#ifdef CONFIG_X86_64
int default_cpu_present_to_apicid(int mps_cpu)
{
	return __default_cpu_present_to_apicid(mps_cpu);
}

T
Thomas Gleixner 已提交
140
int default_check_phys_apicid_present(int phys_apicid)
I
Ingo Molnar 已提交
141
{
T
Thomas Gleixner 已提交
142
	return __default_check_phys_apicid_present(phys_apicid);
I
Ingo Molnar 已提交
143 144 145
}
#endif

146 147 148 149 150 151
#ifndef CONFIG_DEBUG_BOOT_PARAMS
struct boot_params __initdata boot_params;
#else
struct boot_params boot_params;
#endif

L
Linus Torvalds 已提交
152 153 154
/*
 * Machine setup..
 */
155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175
static struct resource data_resource = {
	.name	= "Kernel data",
	.start	= 0,
	.end	= 0,
	.flags	= IORESOURCE_BUSY | IORESOURCE_MEM
};

static struct resource code_resource = {
	.name	= "Kernel code",
	.start	= 0,
	.end	= 0,
	.flags	= IORESOURCE_BUSY | IORESOURCE_MEM
};

static struct resource bss_resource = {
	.name	= "Kernel bss",
	.start	= 0,
	.end	= 0,
	.flags	= IORESOURCE_BUSY | IORESOURCE_MEM
};

176 177

#ifdef CONFIG_X86_32
L
Linus Torvalds 已提交
178
/* cpu data as detected by the assembly code in head.S */
179
struct cpuinfo_x86 new_cpu_data __cpuinitdata = {0, 0, 0, 0, -1, 1, 0, 0, -1};
L
Linus Torvalds 已提交
180
/* common cpu data for all cpus */
181
struct cpuinfo_x86 boot_cpu_data __read_mostly = {0, 0, 0, 0, -1, 1, 0, 0, -1};
182
EXPORT_SYMBOL(boot_cpu_data);
L
Linus Torvalds 已提交
183

184 185
unsigned int def_to_bigsmp;

L
Linus Torvalds 已提交
186 187 188 189 190
/* for MCA, but anyone else can use it if they want */
unsigned int machine_id;
unsigned int machine_submodel_id;
unsigned int BIOS_revision;

191 192 193 194 195 196 197 198 199 200 201 202
struct apm_info apm_info;
EXPORT_SYMBOL(apm_info);

#if defined(CONFIG_X86_SPEEDSTEP_SMI) || \
	defined(CONFIG_X86_SPEEDSTEP_SMI_MODULE)
struct ist_info ist_info;
EXPORT_SYMBOL(ist_info);
#else
struct ist_info ist_info;
#endif

#else
203 204 205
struct cpuinfo_x86 boot_cpu_data __read_mostly = {
	.x86_phys_bits = MAX_PHYSMEM_BITS,
};
206 207 208 209 210 211 212 213 214 215
EXPORT_SYMBOL(boot_cpu_data);
#endif


#if !defined(CONFIG_X86_PAE) || defined(CONFIG_X86_64)
unsigned long mmu_cr4_features;
#else
unsigned long mmu_cr4_features = X86_CR4_PAE;
#endif

216 217
/* Boot loader ID and version as integers, for the benefit of proc_dointvec */
int bootloader_type, bootloader_version;
L
Linus Torvalds 已提交
218 219 220 221 222

/*
 * Setup options
 */
struct screen_info screen_info;
223
EXPORT_SYMBOL(screen_info);
L
Linus Torvalds 已提交
224
struct edid_info edid_info;
225
EXPORT_SYMBOL_GPL(edid_info);
L
Linus Torvalds 已提交
226 227 228

extern int root_mountflags;

229
unsigned long saved_video_mode;
L
Linus Torvalds 已提交
230

231
#define RAMDISK_IMAGE_START_MASK	0x07FF
L
Linus Torvalds 已提交
232
#define RAMDISK_PROMPT_FLAG		0x8000
233
#define RAMDISK_LOAD_FLAG		0x4000
L
Linus Torvalds 已提交
234

235
static char __initdata command_line[COMMAND_LINE_SIZE];
236 237 238
#ifdef CONFIG_CMDLINE_BOOL
static char __initdata builtin_cmdline[COMMAND_LINE_SIZE] = CONFIG_CMDLINE;
#endif
L
Linus Torvalds 已提交
239 240 241 242 243 244 245 246 247 248 249

#if defined(CONFIG_EDD) || defined(CONFIG_EDD_MODULE)
struct edd edd;
#ifdef CONFIG_EDD_MODULE
EXPORT_SYMBOL(edd);
#endif
/**
 * copy_edd() - Copy the BIOS EDD information
 *              from boot_params into a safe place.
 *
 */
250
static inline void __init copy_edd(void)
L
Linus Torvalds 已提交
251
{
252 253 254 255 256
     memcpy(edd.mbr_signature, boot_params.edd_mbr_sig_buffer,
	    sizeof(edd.mbr_signature));
     memcpy(edd.edd_info, boot_params.eddbuf, sizeof(edd.edd_info));
     edd.mbr_signature_nr = boot_params.edd_mbr_sig_buf_entries;
     edd.edd_info_nr = boot_params.eddbuf_entries;
L
Linus Torvalds 已提交
257 258
}
#else
259
static inline void __init copy_edd(void)
L
Linus Torvalds 已提交
260 261 262 263
{
}
#endif

264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282
void * __init extend_brk(size_t size, size_t align)
{
	size_t mask = align - 1;
	void *ret;

	BUG_ON(_brk_start == 0);
	BUG_ON(align & mask);

	_brk_end = (_brk_end + mask) & ~mask;
	BUG_ON((char *)(_brk_end + size) > __brk_limit);

	ret = (void *)_brk_end;
	_brk_end += size;

	memset(ret, 0, size);

	return ret;
}

283 284 285 286 287 288 289 290 291 292 293 294
#ifdef CONFIG_X86_64
static void __init init_gbpages(void)
{
	if (direct_gbpages && cpu_has_gbpages)
		printk(KERN_INFO "Using GB pages for direct mapping\n");
	else
		direct_gbpages = 0;
}
#else
static inline void init_gbpages(void)
{
}
295
static void __init cleanup_highmap(void)
296 297
{
}
298 299
#endif

300 301 302
static void __init reserve_brk(void)
{
	if (_brk_end > _brk_start)
303 304
		memblock_reserve(__pa(_brk_start),
				 __pa(_brk_end) - __pa(_brk_start));
305 306 307 308 309 310

	/* Mark brk area as locked down and no longer taking any
	   new allocations */
	_brk_start = 0;
}

311 312
#ifdef CONFIG_BLK_DEV_INITRD

313 314
#define MAX_MAP_CHUNK	(NR_FIX_BTMAPS << PAGE_SHIFT)
static void __init relocate_initrd(void)
315
{
316
	/* Assume only end is not page aligned */
317 318
	u64 ramdisk_image = boot_params.hdr.ramdisk_image;
	u64 ramdisk_size  = boot_params.hdr.ramdisk_size;
319
	u64 area_size     = PAGE_ALIGN(ramdisk_size);
320
	u64 end_of_lowmem = max_low_pfn_mapped << PAGE_SHIFT;
321
	u64 ramdisk_here;
322 323
	unsigned long slop, clen, mapaddr;
	char *p, *q;
324 325

	/* We need to move the initrd down into lowmem */
326
	ramdisk_here = memblock_find_in_range(0, end_of_lowmem, area_size,
327
					 PAGE_SIZE);
328

T
Tejun Heo 已提交
329
	if (!ramdisk_here)
330 331 332
		panic("Cannot find place for new RAMDISK of size %lld\n",
			 ramdisk_size);

333 334
	/* Note: this includes all the lowmem currently occupied by
	   the initrd, we rely on that fact to keep the data intact. */
335
	memblock_reserve(ramdisk_here, area_size);
336 337
	initrd_start = ramdisk_here + PAGE_OFFSET;
	initrd_end   = initrd_start + ramdisk_size;
338 339
	printk(KERN_INFO "Allocated new RAMDISK: [mem %#010llx-%#010llx]\n",
			 ramdisk_here, ramdisk_here + ramdisk_size - 1);
340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359

	q = (char *)initrd_start;

	/* Copy any lowmem portion of the initrd */
	if (ramdisk_image < end_of_lowmem) {
		clen = end_of_lowmem - ramdisk_image;
		p = (char *)__va(ramdisk_image);
		memcpy(q, p, clen);
		q += clen;
		ramdisk_image += clen;
		ramdisk_size  -= clen;
	}

	/* Copy the highmem portion of the initrd */
	while (ramdisk_size) {
		slop = ramdisk_image & ~PAGE_MASK;
		clen = ramdisk_size;
		if (clen > MAX_MAP_CHUNK-slop)
			clen = MAX_MAP_CHUNK-slop;
		mapaddr = ramdisk_image & PAGE_MASK;
360
		p = early_memremap(mapaddr, clen+slop);
361
		memcpy(q, p+slop, clen);
362
		early_iounmap(p, clen+slop);
363 364 365 366
		q += clen;
		ramdisk_image += clen;
		ramdisk_size  -= clen;
	}
367
	/* high pages is not converted by early_res_to_bootmem */
368 369
	ramdisk_image = boot_params.hdr.ramdisk_image;
	ramdisk_size  = boot_params.hdr.ramdisk_size;
370 371
	printk(KERN_INFO "Move RAMDISK from [mem %#010llx-%#010llx] to"
		" [mem %#010llx-%#010llx]\n",
372 373
		ramdisk_image, ramdisk_image + ramdisk_size - 1,
		ramdisk_here, ramdisk_here + ramdisk_size - 1);
374
}
Y
Yinghai Lu 已提交
375

376 377
static void __init reserve_initrd(void)
{
378
	/* Assume only end is not page aligned */
379 380
	u64 ramdisk_image = boot_params.hdr.ramdisk_image;
	u64 ramdisk_size  = boot_params.hdr.ramdisk_size;
381
	u64 ramdisk_end   = PAGE_ALIGN(ramdisk_image + ramdisk_size);
382
	u64 end_of_lowmem = max_low_pfn_mapped << PAGE_SHIFT;
383 384 385 386 387 388 389 390

	if (!boot_params.hdr.type_of_loader ||
	    !ramdisk_image || !ramdisk_size)
		return;		/* No initrd provided by bootloader */

	initrd_start = 0;

	if (ramdisk_size >= (end_of_lowmem>>1)) {
391 392 393
		panic("initrd too large to handle, "
		       "disabling initrd (%lld needed, %lld available)\n",
		       ramdisk_size, end_of_lowmem>>1);
394 395
	}

396 397
	printk(KERN_INFO "RAMDISK: [mem %#010llx-%#010llx]\n", ramdisk_image,
			ramdisk_end - 1);
398 399 400 401 402 403 404 405 406 407 408 409 410 411


	if (ramdisk_end <= end_of_lowmem) {
		/* All in lowmem, easy case */
		/*
		 * don't need to reserve again, already reserved early
		 * in i386_start_kernel
		 */
		initrd_start = ramdisk_image + PAGE_OFFSET;
		initrd_end = initrd_start + ramdisk_size;
		return;
	}

	relocate_initrd();
412

413
	memblock_free(ramdisk_image, ramdisk_end - ramdisk_image);
414
}
Y
Yinghai Lu 已提交
415
#else
416
static void __init reserve_initrd(void)
Y
Yinghai Lu 已提交
417 418
{
}
419 420
#endif /* CONFIG_BLK_DEV_INITRD */

421
static void __init parse_setup_data(void)
422 423 424 425 426 427 428 429
{
	struct setup_data *data;
	u64 pa_data;

	if (boot_params.hdr.version < 0x0209)
		return;
	pa_data = boot_params.hdr.setup_data;
	while (pa_data) {
430 431 432 433 434 435 436 437 438 439 440 441
		u32 data_len, map_len;

		map_len = max(PAGE_SIZE - (pa_data & ~PAGE_MASK),
			      (u64)sizeof(struct setup_data));
		data = early_memremap(pa_data, map_len);
		data_len = data->len + sizeof(struct setup_data);
		if (data_len > map_len) {
			early_iounmap(data, map_len);
			data = early_memremap(pa_data, data_len);
			map_len = data_len;
		}

442 443
		switch (data->type) {
		case SETUP_E820_EXT:
444
			parse_e820_ext(data);
445
			break;
446 447
		case SETUP_DTB:
			add_dtb(pa_data);
448 449 450 451 452
			break;
		default:
			break;
		}
		pa_data = data->next;
453
		early_iounmap(data, map_len);
454 455 456
	}
}

457
static void __init e820_reserve_setup_data(void)
458 459 460
{
	struct setup_data *data;
	u64 pa_data;
461
	int found = 0;
462 463 464 465 466

	if (boot_params.hdr.version < 0x0209)
		return;
	pa_data = boot_params.hdr.setup_data;
	while (pa_data) {
467
		data = early_memremap(pa_data, sizeof(*data));
468 469
		e820_update_range(pa_data, sizeof(*data)+data->len,
			 E820_RAM, E820_RESERVED_KERN);
470
		found = 1;
471 472 473
		pa_data = data->next;
		early_iounmap(data, sizeof(*data));
	}
474 475 476
	if (!found)
		return;

477
	sanitize_e820_map(e820.map, ARRAY_SIZE(e820.map), &e820.nr_map);
478
	memcpy(&e820_saved, &e820, sizeof(struct e820map));
479 480 481 482
	printk(KERN_INFO "extended physical RAM map:\n");
	e820_print_map("reserve setup_data");
}

483
static void __init memblock_x86_reserve_range_setup_data(void)
484 485 486 487 488 489 490 491
{
	struct setup_data *data;
	u64 pa_data;

	if (boot_params.hdr.version < 0x0209)
		return;
	pa_data = boot_params.hdr.setup_data;
	while (pa_data) {
492
		data = early_memremap(pa_data, sizeof(*data));
493
		memblock_reserve(pa_data, sizeof(*data) + data->len);
494 495 496 497 498
		pa_data = data->next;
		early_iounmap(data, sizeof(*data));
	}
}

499 500 501 502 503
/*
 * --------- Crashkernel reservation ------------------------------
 */

#ifdef CONFIG_KEXEC
504

505 506 507 508 509 510 511 512 513 514 515 516
/*
 * Keep the crash kernel below this limit.  On 32 bits earlier kernels
 * would limit the kernel to the low 512 MiB due to mapping restrictions.
 * On 64 bits, kexec-tools currently limits us to 896 MiB; increase this
 * limit once kexec-tools are fixed.
 */
#ifdef CONFIG_X86_32
# define CRASH_KERNEL_ADDR_MAX	(512 << 20)
#else
# define CRASH_KERNEL_ADDR_MAX	(896 << 20)
#endif

517
static void __init reserve_crashkernel(void)
518 519 520 521 522
{
	unsigned long long total_mem;
	unsigned long long crash_size, crash_base;
	int ret;

523
	total_mem = memblock_phys_mem_size();
524 525 526

	ret = parse_crashkernel(boot_command_line, total_mem,
			&crash_size, &crash_base);
527 528 529 530 531
	if (ret != 0 || crash_size <= 0)
		return;

	/* 0 means: find the address automatically */
	if (crash_base <= 0) {
532 533
		const unsigned long long alignment = 16<<20;	/* 16M */

534
		/*
535
		 *  kexec want bzImage is below CRASH_KERNEL_ADDR_MAX
536 537
		 */
		crash_base = memblock_find_in_range(alignment,
538
			       CRASH_KERNEL_ADDR_MAX, crash_size, alignment);
539

T
Tejun Heo 已提交
540
		if (!crash_base) {
541
			pr_info("crashkernel reservation failed - No suitable area found.\n");
542 543
			return;
		}
544
	} else {
545 546
		unsigned long long start;

547 548
		start = memblock_find_in_range(crash_base,
				 crash_base + crash_size, crash_size, 1<<20);
549 550
		if (start != crash_base) {
			pr_info("crashkernel reservation failed - memory is in use.\n");
551 552
			return;
		}
553
	}
554
	memblock_reserve(crash_base, crash_size);
555

556 557 558 559 560
	printk(KERN_INFO "Reserving %ldMB of memory at %ldMB "
			"for crashkernel (System RAM: %ldMB)\n",
			(unsigned long)(crash_size >> 20),
			(unsigned long)(crash_base >> 20),
			(unsigned long)(total_mem >> 20));
561

562 563 564
	crashk_res.start = crash_base;
	crashk_res.end   = crash_base + crash_size - 1;
	insert_resource(&iomem_resource, &crashk_res);
565 566
}
#else
567
static void __init reserve_crashkernel(void)
568 569 570 571
{
}
#endif

572 573 574 575 576 577 578 579 580 581 582 583 584 585 586 587 588 589 590 591 592 593 594
static struct resource standard_io_resources[] = {
	{ .name = "dma1", .start = 0x00, .end = 0x1f,
		.flags = IORESOURCE_BUSY | IORESOURCE_IO },
	{ .name = "pic1", .start = 0x20, .end = 0x21,
		.flags = IORESOURCE_BUSY | IORESOURCE_IO },
	{ .name = "timer0", .start = 0x40, .end = 0x43,
		.flags = IORESOURCE_BUSY | IORESOURCE_IO },
	{ .name = "timer1", .start = 0x50, .end = 0x53,
		.flags = IORESOURCE_BUSY | IORESOURCE_IO },
	{ .name = "keyboard", .start = 0x60, .end = 0x60,
		.flags = IORESOURCE_BUSY | IORESOURCE_IO },
	{ .name = "keyboard", .start = 0x64, .end = 0x64,
		.flags = IORESOURCE_BUSY | IORESOURCE_IO },
	{ .name = "dma page reg", .start = 0x80, .end = 0x8f,
		.flags = IORESOURCE_BUSY | IORESOURCE_IO },
	{ .name = "pic2", .start = 0xa0, .end = 0xa1,
		.flags = IORESOURCE_BUSY | IORESOURCE_IO },
	{ .name = "dma2", .start = 0xc0, .end = 0xdf,
		.flags = IORESOURCE_BUSY | IORESOURCE_IO },
	{ .name = "fpu", .start = 0xf0, .end = 0xff,
		.flags = IORESOURCE_BUSY | IORESOURCE_IO }
};

595
void __init reserve_standard_io_resources(void)
596 597 598 599 600 601 602 603 604
{
	int i;

	/* request I/O space for devices used on all i[345]86 PCs */
	for (i = 0; i < ARRAY_SIZE(standard_io_resources); i++)
		request_resource(&ioport_resource, &standard_io_resources[i]);

}

605 606 607 608 609 610 611
static __init void reserve_ibft_region(void)
{
	unsigned long addr, size = 0;

	addr = find_ibft_region(&size);

	if (size)
612
		memblock_reserve(addr, size);
613 614
}

615
static unsigned reserve_low = CONFIG_X86_RESERVE_LOW << 10;
616

617 618 619 620 621 622
static void __init trim_bios_range(void)
{
	/*
	 * A special case is the first 4Kb of memory;
	 * This is a BIOS owned area, not kernel ram, but generally
	 * not listed as such in the E820 table.
623 624 625
	 *
	 * This typically reserves additional memory (64KiB by default)
	 * since some BIOSes are known to corrupt low memory.  See the
626
	 * Kconfig help text for X86_RESERVE_LOW.
627
	 */
628
	e820_update_range(0, ALIGN(reserve_low, PAGE_SIZE),
629 630
			  E820_RAM, E820_RESERVED);

631 632 633 634 635 636 637 638 639
	/*
	 * special case: Some BIOSen report the PC BIOS
	 * area (640->1Mb) as ram even though it is not.
	 * take them out.
	 */
	e820_remove_range(BIOS_BEGIN, BIOS_END - BIOS_BEGIN, E820_RAM, 1);
	sanitize_e820_map(e820.map, ARRAY_SIZE(e820.map), &e820.nr_map);
}

640 641 642 643 644 645 646 647 648 649 650 651 652 653 654 655 656 657 658 659 660 661
static int __init parse_reservelow(char *p)
{
	unsigned long long size;

	if (!p)
		return -EINVAL;

	size = memparse(p, &p);

	if (size < 4096)
		size = 4096;

	if (size > 640*1024)
		size = 640*1024;

	reserve_low = size;

	return 0;
}

early_param("reservelow", parse_reservelow);

L
Linus Torvalds 已提交
662 663 664 665 666 667 668
/*
 * Determine if we were loaded by an EFI loader.  If so, then we have also been
 * passed the efi memmap, systab, etc., so we should use these data structures
 * for initialization.  Note, the efi init code path is determined by the
 * global efi_enabled. This allows the same kernel image to be used on existing
 * systems (with a traditional BIOS) as well as on EFI systems.
 */
669 670 671 672 673 674
/*
 * setup_arch - architecture-specific boot-time initializations
 *
 * Note: On x86_64, fixmaps are ready for use even before this is called.
 */

L
Linus Torvalds 已提交
675 676
void __init setup_arch(char **cmdline_p)
{
677
#ifdef CONFIG_X86_32
L
Linus Torvalds 已提交
678
	memcpy(&boot_cpu_data, &new_cpu_data, sizeof(new_cpu_data));
I
Ingo Molnar 已提交
679
	visws_early_detect();
680 681 682 683 684 685 686 687 688 689 690

	/*
	 * copy kernel address range established so far and switch
	 * to the proper swapper page table
	 */
	clone_pgd_range(swapper_pg_dir     + KERNEL_PGD_BOUNDARY,
			initial_page_table + KERNEL_PGD_BOUNDARY,
			KERNEL_PGD_PTRS);

	load_cr3(swapper_pg_dir);
	__flush_tlb_all();
691 692 693
#else
	printk(KERN_INFO "Command line: %s\n", boot_command_line);
#endif
L
Linus Torvalds 已提交
694

695 696 697 698
	/*
	 * If we have OLPC OFW, we might end up relocating the fixmap due to
	 * reserve_top(), so do this before touching the ioremap area.
	 */
699 700
	olpc_ofw_detect();

701
	early_trap_init();
702
	early_cpu_init();
703 704
	early_ioremap_init();

705 706
	setup_olpc_ofw_pgd();

707 708 709
	ROOT_DEV = old_decode_dev(boot_params.hdr.root_dev);
	screen_info = boot_params.screen_info;
	edid_info = boot_params.edid_info;
710
#ifdef CONFIG_X86_32
711 712
	apm_info.bios = boot_params.apm_bios_info;
	ist_info = boot_params.ist_info;
713
	if (boot_params.sys_desc_table.length != 0) {
714 715 716
		machine_id = boot_params.sys_desc_table.table[0];
		machine_submodel_id = boot_params.sys_desc_table.table[1];
		BIOS_revision = boot_params.sys_desc_table.table[2];
L
Linus Torvalds 已提交
717
	}
718 719
#endif
	saved_video_mode = boot_params.hdr.vid_mode;
720
	bootloader_type = boot_params.hdr.type_of_loader;
721 722 723 724 725 726
	if ((bootloader_type >> 4) == 0xe) {
		bootloader_type &= 0xf;
		bootloader_type |= (boot_params.hdr.ext_loader_type+0x10) << 4;
	}
	bootloader_version  = bootloader_type & 0xf;
	bootloader_version |= boot_params.hdr.ext_loader_ver << 4;
L
Linus Torvalds 已提交
727 728

#ifdef CONFIG_BLK_DEV_RAM
729 730 731
	rd_image_start = boot_params.hdr.ram_size & RAMDISK_IMAGE_START_MASK;
	rd_prompt = ((boot_params.hdr.ram_size & RAMDISK_PROMPT_FLAG) != 0);
	rd_doload = ((boot_params.hdr.ram_size & RAMDISK_LOAD_FLAG) != 0);
L
Linus Torvalds 已提交
732
#endif
733 734
#ifdef CONFIG_EFI
	if (!strncmp((char *)&boot_params.efi_info.efi_loader_signature,
735
		     "EL32", 4)) {
736
		efi_enabled = 1;
737 738 739 740 741
		efi_64bit = false;
	} else if (!strncmp((char *)&boot_params.efi_info.efi_loader_signature,
		     "EL64", 4)) {
		efi_enabled = 1;
		efi_64bit = true;
742
	}
743 744
	if (efi_enabled && efi_memblock_x86_reserve_range())
		efi_enabled = 0;
745 746
#endif

747
	x86_init.oem.arch_setup();
748

749
	iomem_resource.end = (1ULL << boot_cpu_data.x86_phys_bits) - 1;
750
	setup_memory_map();
751
	parse_setup_data();
752 753
	/* update the e820_saved too */
	e820_reserve_setup_data();
754

L
Linus Torvalds 已提交
755 756
	copy_edd();

757
	if (!boot_params.hdr.root_flags)
L
Linus Torvalds 已提交
758 759 760 761
		root_mountflags &= ~MS_RDONLY;
	init_mm.start_code = (unsigned long) _text;
	init_mm.end_code = (unsigned long) _etext;
	init_mm.end_data = (unsigned long) _edata;
762
	init_mm.brk = _brk_end;
L
Linus Torvalds 已提交
763 764 765 766 767

	code_resource.start = virt_to_phys(_text);
	code_resource.end = virt_to_phys(_etext)-1;
	data_resource.start = virt_to_phys(_etext);
	data_resource.end = virt_to_phys(_edata)-1;
B
Bernhard Walle 已提交
768 769
	bss_resource.start = virt_to_phys(&__bss_start);
	bss_resource.end = virt_to_phys(&__bss_stop)-1;
L
Linus Torvalds 已提交
770

771 772 773 774 775 776 777 778 779 780 781 782 783
#ifdef CONFIG_CMDLINE_BOOL
#ifdef CONFIG_CMDLINE_OVERRIDE
	strlcpy(boot_command_line, builtin_cmdline, COMMAND_LINE_SIZE);
#else
	if (builtin_cmdline[0]) {
		/* append boot loader cmdline to builtin */
		strlcat(builtin_cmdline, " ", COMMAND_LINE_SIZE);
		strlcat(builtin_cmdline, boot_command_line, COMMAND_LINE_SIZE);
		strlcpy(boot_command_line, builtin_cmdline, COMMAND_LINE_SIZE);
	}
#endif
#endif

784 785 786 787
	strlcpy(command_line, boot_command_line, COMMAND_LINE_SIZE);
	*cmdline_p = command_line;

	/*
788 789 790 791 792
	 * x86_configure_nx() is called before parse_early_param() to detect
	 * whether hardware doesn't support NX (so that the early EHCI debug
	 * console setup can safely call set_fixmap()). It may then be called
	 * again from within noexec_setup() during parsing early parameters
	 * to honor the respective command line option.
793
	 */
794
	x86_configure_nx();
795 796 797

	parse_early_param();

798
	x86_report_nx();
799

800
	/* after early param, so could get panic from serial */
801
	memblock_x86_reserve_range_setup_data();
802

803
	if (acpi_mps_check()) {
I
Ingo Molnar 已提交
804
#ifdef CONFIG_X86_LOCAL_APIC
805
		disable_apic = 1;
I
Ingo Molnar 已提交
806
#endif
807
		setup_clear_cpu_cap(X86_FEATURE_APIC);
808 809
	}

810 811 812 813 814
#ifdef CONFIG_PCI
	if (pci_early_dump_regs)
		early_dump_pci_devices();
#endif

815
	finish_e820_parsing();
816

B
Brian Maly 已提交
817 818 819
	if (efi_enabled)
		efi_init();

820 821
	dmi_scan_machine();

822 823 824 825
	/*
	 * VMware detection requires dmi to be available, so this
	 * needs to be done after dmi_scan_machine, for the BP.
	 */
826
	init_hypervisor_platform();
827

828
	x86_init.resources.probe_roms();
829 830 831 832 833 834

	/* after parse_early_param, so could debug it */
	insert_resource(&iomem_resource, &code_resource);
	insert_resource(&iomem_resource, &data_resource);
	insert_resource(&iomem_resource, &bss_resource);

835 836 837 838 839 840 841 842 843 844 845 846 847 848
	/*
	 * Complain if .text .data and .bss are not marked as E820_RAM and
	 * attempt to fix it by adding the range. We may have a confused BIOS,
	 * or the user may have incorrectly supplied it via memmap=exactmap. If
	 * we really are running on top non-RAM, we will crash later anyways.
	 */
	if (!e820_all_mapped(code_resource.start, __pa(__brk_limit), E820_RAM)) {
		pr_warn(".text .data .bss are not marked as E820_RAM!\n");

		e820_add_region(code_resource.start,
				__pa(__brk_limit) - code_resource.start + 1,
				E820_RAM);
	}

849
	trim_bios_range();
850
#ifdef CONFIG_X86_32
Y
Yinghai Lu 已提交
851 852 853 854 855 856 857
	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);
		printk(KERN_INFO "fixed physical RAM map:\n");
		e820_print_map("bad_ppro");
	}
858 859 860
#else
	early_gart_iommu_check();
#endif
Y
Yinghai Lu 已提交
861

862 863 864 865
	/*
	 * partially used pages are not usable - thus
	 * we are rounding upwards:
	 */
866
	max_pfn = e820_end_of_ram_pfn();
867

868 869
	/* update e820 for memory not covered by WB MTRRs */
	mtrr_bp_init();
870
	if (mtrr_trim_uncached_memory(max_pfn))
871
		max_pfn = e820_end_of_ram_pfn();
872

873
#ifdef CONFIG_X86_32
874
	/* max_low_pfn get updated here */
875
	find_low_pfn_range();
876 877 878
#else
	num_physpages = max_pfn;

Y
Yinghai Lu 已提交
879
	check_x2apic();
880 881 882

	/* How many end-of-memory variables you have, grandma! */
	/* need this before calling reserve_initrd */
883 884 885 886 887
	if (max_pfn > (1UL<<(32 - PAGE_SHIFT)))
		max_low_pfn = e820_end_of_low_ram_pfn();
	else
		max_low_pfn = max_pfn;

888
	high_memory = (void *)__va(max_pfn * PAGE_SIZE - 1) + 1;
889 890
#endif

891 892 893 894 895
	/*
	 * Find and reserve possible boot-time SMP configuration:
	 */
	find_smp_config();

896 897
	reserve_ibft_region();

898 899 900 901 902 903 904
	/*
	 * Need to conclude brk, before memblock_x86_fill()
	 *  it could use memblock_find_in_range, could overlap with
	 *  brk area.
	 */
	reserve_brk();

905 906
	cleanup_highmap();

907
	memblock.current_limit = ISA_END_ADDRESS;
908 909
	memblock_x86_fill();

910 911 912 913 914 915 916
	/*
	 * The EFI specification says that boot service code won't be called
	 * after ExitBootServices(). This is, in fact, a lie.
	 */
	if (efi_enabled)
		efi_reserve_boot_services();

917 918 919 920 921 922 923
	/* preallocate 4k for mptable mpc */
	early_reserve_e820_mpc_new();

#ifdef CONFIG_X86_CHECK_BIOS_CORRUPTION
	setup_bios_corruption_check();
#endif

924 925
	printk(KERN_DEBUG "initial memory mapped: [mem 0x00000000-%#010lx]\n",
			(max_pfn_mapped<<PAGE_SHIFT) - 1);
926

927
	setup_real_mode();
928

929 930
	init_gbpages();

931
	init_mem_mapping();
932

933
	memblock.current_limit = get_max_mapped();
934
	dma_contiguous_reserve(0);
935

936 937 938 939 940 941 942 943
	/*
	 * NOTE: On x86-32, only from this point on, fixmaps are ready for use.
	 */

#ifdef CONFIG_PROVIDE_OHCI1394_DMA_INIT
	if (init_ohci1394_dma_early)
		init_ohci1394_dma_on_all_controllers();
#endif
944 945
	/* Allocate bigger log buffer */
	setup_log_buf(1);
946

947 948
	reserve_initrd();

949 950
	reserve_crashkernel();

951 952
	vsmp_init();

953 954 955 956 957 958 959
	io_delay_init();

	/*
	 * Parse the ACPI tables for possible boot-time SMP configuration.
	 */
	acpi_boot_table_init();

960 961
	early_acpi_boot_init();

962
	initmem_init();
963
	memblock_find_dma_reserve();
964

965
#ifdef CONFIG_KVM_GUEST
966 967 968
	kvmclock_init();
#endif

969
	x86_init.paging.pagetable_init();
970

971 972 973
	if (boot_cpu_data.cpuid_level >= 0) {
		/* A CPU has %cr4 if and only if it has CPUID */
		mmu_cr4_features = read_cr4();
974 975
		if (trampoline_cr4_features)
			*trampoline_cr4_features = mmu_cr4_features;
976 977
	}

978 979 980 981 982 983
#ifdef CONFIG_X86_32
	/* sync back kernel address range */
	clone_pgd_range(initial_page_table + KERNEL_PGD_BOUNDARY,
			swapper_pg_dir     + KERNEL_PGD_BOUNDARY,
			KERNEL_PGD_PTRS);
#endif
984

985 986
	tboot_probe();

987 988 989 990
#ifdef CONFIG_X86_64
	map_vsyscall();
#endif

991
	generic_apic_probe();
L
Linus Torvalds 已提交
992

993
	early_quirks();
994

995 996 997
	/*
	 * Read APIC and some other early information from ACPI tables.
	 */
L
Linus Torvalds 已提交
998
	acpi_boot_init();
F
Feng Tang 已提交
999
	sfi_init();
1000
	x86_dtb_init();
1001

1002 1003 1004
	/*
	 * get boot-time SMP configuration:
	 */
1005 1006
	if (smp_found_config)
		get_smp_config();
1007

1008
	prefill_possible_map();
Y
Yinghai Lu 已提交
1009

1010 1011
	init_cpu_to_node();

1012
	init_apic_mappings();
I
Ingo Molnar 已提交
1013 1014
	if (x86_io_apic_ops.init)
		x86_io_apic_ops.init();
1015

1016
	kvm_guest_init();
L
Linus Torvalds 已提交
1017

1018
	e820_reserve_resources();
1019
	e820_mark_nosave_regions(max_low_pfn);
L
Linus Torvalds 已提交
1020

1021
	x86_init.resources.reserve_resources();
1022 1023 1024

	e820_setup_gap();

L
Linus Torvalds 已提交
1025 1026 1027 1028 1029 1030 1031 1032
#ifdef CONFIG_VT
#if defined(CONFIG_VGA_CONSOLE)
	if (!efi_enabled || (efi_mem_type(0xa0000) != EFI_CONVENTIONAL_MEMORY))
		conswitchp = &vga_con;
#elif defined(CONFIG_DUMMY_CONSOLE)
	conswitchp = &dummy_con;
#endif
#endif
1033
	x86_init.oem.banner();
1034

1035 1036
	x86_init.timers.wallclock_init();

1037
	mcheck_init();
1038

1039
	arch_init_ideal_nops();
1040 1041

	register_refined_jiffies(CLOCK_TICK_RATE);
1042 1043 1044 1045 1046 1047 1048 1049 1050 1051 1052 1053

#ifdef CONFIG_EFI
	/* Once setup is done above, disable efi_enabled on mismatched
	 * firmware/kernel archtectures since there is no support for
	 * runtime services.
	 */
	if (efi_enabled && IS_ENABLED(CONFIG_X86_64) != efi_64bit) {
		pr_info("efi: Setup done, disabling due to 32/64-bit mismatch\n");
		efi_unmap_memmap();
		efi_enabled = 0;
	}
#endif
L
Linus Torvalds 已提交
1054
}
1055

1056 1057
#ifdef CONFIG_X86_32

1058 1059 1060 1061 1062
static struct resource video_ram_resource = {
	.name	= "Video RAM area",
	.start	= 0xa0000,
	.end	= 0xbffff,
	.flags	= IORESOURCE_BUSY | IORESOURCE_MEM
1063 1064
};

1065
void __init i386_reserve_resources(void)
1066
{
1067 1068
	request_resource(&iomem_resource, &video_ram_resource);
	reserve_standard_io_resources();
1069 1070 1071
}

#endif /* CONFIG_X86_32 */