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 43
#include <linux/seq_file.h>
#include <linux/console.h>
#include <linux/mca.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>
44
#include <linux/iscsi_ibft.h>
L
Linus Torvalds 已提交
45
#include <linux/nodemask.h>
46
#include <linux/kexec.h>
47
#include <linux/dmi.h>
D
Dave Hansen 已提交
48
#include <linux/pfn.h>
49
#include <linux/pci.h>
50
#include <asm/pci-direct.h>
51
#include <linux/init_ohci1394_dma.h>
52
#include <linux/kvm_para.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

L
Linus Torvalds 已提交
72
#include <video/edid.h>
73

74
#include <asm/mtrr.h>
75
#include <asm/apic.h>
76
#include <asm/trampoline.h>
L
Linus Torvalds 已提交
77 78 79
#include <asm/e820.h>
#include <asm/mpspec.h>
#include <asm/setup.h>
Y
Yinghai Lu 已提交
80
#include <asm/efi.h>
I
Ingo Molnar 已提交
81 82
#include <asm/timer.h>
#include <asm/i8259.h>
L
Linus Torvalds 已提交
83
#include <asm/sections.h>
84
#include <asm/dmi.h>
L
Linus Torvalds 已提交
85 86
#include <asm/io_apic.h>
#include <asm/ist.h>
87
#include <asm/setup_arch.h>
88
#include <asm/bios_ebda.h>
B
Bernhard Walle 已提交
89
#include <asm/cacheflush.h>
90
#include <asm/processor.h>
Y
Yinghai Lu 已提交
91
#include <asm/bugs.h>
L
Linus Torvalds 已提交
92

93 94
#include <asm/system.h>
#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

117 118 119 120 121 122 123 124
/*
 * 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;

125
#ifdef CONFIG_DMI
126
RESERVE_BRK(dmi_alloc, 65536);
127
#endif
128

I
Ingo Molnar 已提交
129

130 131 132
static __initdata unsigned long _brk_start = (unsigned long)__brk_base;
unsigned long _brk_end = (unsigned long)__brk_base;

I
Ingo Molnar 已提交
133 134 135 136 137 138
#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
139
int default_check_phys_apicid_present(int phys_apicid)
I
Ingo Molnar 已提交
140
{
T
Thomas Gleixner 已提交
141
	return __default_check_phys_apicid_present(phys_apicid);
I
Ingo Molnar 已提交
142 143 144
}
#endif

145 146 147 148 149 150
#ifndef CONFIG_DEBUG_BOOT_PARAMS
struct boot_params __initdata boot_params;
#else
struct boot_params boot_params;
#endif

L
Linus Torvalds 已提交
151 152 153
/*
 * Machine setup..
 */
154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174
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
};

175 176

#ifdef CONFIG_X86_32
L
Linus Torvalds 已提交
177
/* cpu data as detected by the assembly code in head.S */
178
struct cpuinfo_x86 new_cpu_data __cpuinitdata = {0, 0, 0, 0, -1, 1, 0, 0, -1};
L
Linus Torvalds 已提交
179
/* common cpu data for all cpus */
180
struct cpuinfo_x86 boot_cpu_data __read_mostly = {0, 0, 0, 0, -1, 1, 0, 0, -1};
181
EXPORT_SYMBOL(boot_cpu_data);
182 183 184 185 186 187
static void set_mca_bus(int x)
{
#ifdef CONFIG_MCA
	MCA_bus = x;
#endif
}
L
Linus Torvalds 已提交
188

189 190
unsigned int def_to_bigsmp;

L
Linus Torvalds 已提交
191 192 193 194 195
/* for MCA, but anyone else can use it if they want */
unsigned int machine_id;
unsigned int machine_submodel_id;
unsigned int BIOS_revision;

196 197 198 199 200 201 202 203 204 205 206 207
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
208 209 210
struct cpuinfo_x86 boot_cpu_data __read_mostly = {
	.x86_phys_bits = MAX_PHYSMEM_BITS,
};
211 212 213 214 215 216 217 218 219 220
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

221 222
/* Boot loader ID and version as integers, for the benefit of proc_dointvec */
int bootloader_type, bootloader_version;
L
Linus Torvalds 已提交
223 224 225 226 227

/*
 * Setup options
 */
struct screen_info screen_info;
228
EXPORT_SYMBOL(screen_info);
L
Linus Torvalds 已提交
229
struct edid_info edid_info;
230
EXPORT_SYMBOL_GPL(edid_info);
L
Linus Torvalds 已提交
231 232 233

extern int root_mountflags;

234
unsigned long saved_video_mode;
L
Linus Torvalds 已提交
235

236
#define RAMDISK_IMAGE_START_MASK	0x07FF
L
Linus Torvalds 已提交
237
#define RAMDISK_PROMPT_FLAG		0x8000
238
#define RAMDISK_LOAD_FLAG		0x4000
L
Linus Torvalds 已提交
239

240
static char __initdata command_line[COMMAND_LINE_SIZE];
241 242 243
#ifdef CONFIG_CMDLINE_BOOL
static char __initdata builtin_cmdline[COMMAND_LINE_SIZE] = CONFIG_CMDLINE;
#endif
L
Linus Torvalds 已提交
244 245 246 247 248 249 250 251 252 253 254

#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.
 *
 */
255
static inline void __init copy_edd(void)
L
Linus Torvalds 已提交
256
{
257 258 259 260 261
     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
262 263
}
#else
264
static inline void __init copy_edd(void)
L
Linus Torvalds 已提交
265 266 267 268
{
}
#endif

269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287
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;
}

288 289 290 291 292 293 294 295 296 297 298 299 300 301
#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)
{
}
#endif

302 303 304
static void __init reserve_brk(void)
{
	if (_brk_end > _brk_start)
305
		memblock_x86_reserve_range(__pa(_brk_start), __pa(_brk_end), "BRK");
306 307 308 309 310 311

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

312 313
#ifdef CONFIG_BLK_DEV_INITRD

314 315
#define MAX_MAP_CHUNK	(NR_FIX_BTMAPS << PAGE_SHIFT)
static void __init relocate_initrd(void)
316
{
317
	/* Assume only end is not page aligned */
318 319
	u64 ramdisk_image = boot_params.hdr.ramdisk_image;
	u64 ramdisk_size  = boot_params.hdr.ramdisk_size;
320
	u64 area_size     = PAGE_ALIGN(ramdisk_size);
321
	u64 end_of_lowmem = max_low_pfn_mapped << PAGE_SHIFT;
322
	u64 ramdisk_here;
323 324
	unsigned long slop, clen, mapaddr;
	char *p, *q;
325 326

	/* We need to move the initrd down into lowmem */
327
	ramdisk_here = memblock_find_in_range(0, end_of_lowmem, area_size,
328
					 PAGE_SIZE);
329

330
	if (ramdisk_here == MEMBLOCK_ERROR)
331 332 333
		panic("Cannot find place for new RAMDISK of size %lld\n",
			 ramdisk_size);

334 335
	/* Note: this includes all the lowmem currently occupied by
	   the initrd, we rely on that fact to keep the data intact. */
336
	memblock_x86_reserve_range(ramdisk_here, ramdisk_here + area_size, "NEW RAMDISK");
337 338
	initrd_start = ramdisk_here + PAGE_OFFSET;
	initrd_end   = initrd_start + ramdisk_size;
339 340
	printk(KERN_INFO "Allocated new RAMDISK: %08llx - %08llx\n",
			 ramdisk_here, ramdisk_here + ramdisk_size);
341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360

	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;
361
		p = early_memremap(mapaddr, clen+slop);
362
		memcpy(q, p+slop, clen);
363
		early_iounmap(p, clen+slop);
364 365 366 367
		q += clen;
		ramdisk_image += clen;
		ramdisk_size  -= clen;
	}
368
	/* high pages is not converted by early_res_to_bootmem */
369 370
	ramdisk_image = boot_params.hdr.ramdisk_image;
	ramdisk_size  = boot_params.hdr.ramdisk_size;
371 372
	printk(KERN_INFO "Move RAMDISK from %016llx - %016llx to"
		" %08llx - %08llx\n",
373 374
		ramdisk_image, ramdisk_image + ramdisk_size - 1,
		ramdisk_here, ramdisk_here + ramdisk_size - 1);
375
}
Y
Yinghai Lu 已提交
376

377 378
static void __init reserve_initrd(void)
{
379
	/* Assume only end is not page aligned */
380 381
	u64 ramdisk_image = boot_params.hdr.ramdisk_image;
	u64 ramdisk_size  = boot_params.hdr.ramdisk_size;
382
	u64 ramdisk_end   = PAGE_ALIGN(ramdisk_image + ramdisk_size);
383
	u64 end_of_lowmem = max_low_pfn_mapped << PAGE_SHIFT;
384 385 386 387 388 389 390 391

	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)) {
392
		memblock_x86_free_range(ramdisk_image, ramdisk_end);
393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413
		printk(KERN_ERR "initrd too large to handle, "
		       "disabling initrd\n");
		return;
	}

	printk(KERN_INFO "RAMDISK: %08llx - %08llx\n", ramdisk_image,
			ramdisk_end);


	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();
414

415
	memblock_x86_free_range(ramdisk_image, ramdisk_end);
416
}
Y
Yinghai Lu 已提交
417
#else
418
static void __init reserve_initrd(void)
Y
Yinghai Lu 已提交
419 420
{
}
421 422
#endif /* CONFIG_BLK_DEV_INITRD */

423
static void __init parse_setup_data(void)
424 425 426 427 428 429 430 431
{
	struct setup_data *data;
	u64 pa_data;

	if (boot_params.hdr.version < 0x0209)
		return;
	pa_data = boot_params.hdr.setup_data;
	while (pa_data) {
432
		data = early_memremap(pa_data, PAGE_SIZE);
433 434 435 436 437 438 439 440 441 442 443 444
		switch (data->type) {
		case SETUP_E820_EXT:
			parse_e820_ext(data, pa_data);
			break;
		default:
			break;
		}
		pa_data = data->next;
		early_iounmap(data, PAGE_SIZE);
	}
}

445
static void __init e820_reserve_setup_data(void)
446 447 448
{
	struct setup_data *data;
	u64 pa_data;
449
	int found = 0;
450 451 452 453 454

	if (boot_params.hdr.version < 0x0209)
		return;
	pa_data = boot_params.hdr.setup_data;
	while (pa_data) {
455
		data = early_memremap(pa_data, sizeof(*data));
456 457
		e820_update_range(pa_data, sizeof(*data)+data->len,
			 E820_RAM, E820_RESERVED_KERN);
458
		found = 1;
459 460 461
		pa_data = data->next;
		early_iounmap(data, sizeof(*data));
	}
462 463 464
	if (!found)
		return;

465
	sanitize_e820_map(e820.map, ARRAY_SIZE(e820.map), &e820.nr_map);
466
	memcpy(&e820_saved, &e820, sizeof(struct e820map));
467 468 469 470
	printk(KERN_INFO "extended physical RAM map:\n");
	e820_print_map("reserve setup_data");
}

471
static void __init memblock_x86_reserve_range_setup_data(void)
472 473 474 475 476 477 478 479 480
{
	struct setup_data *data;
	u64 pa_data;
	char buf[32];

	if (boot_params.hdr.version < 0x0209)
		return;
	pa_data = boot_params.hdr.setup_data;
	while (pa_data) {
481
		data = early_memremap(pa_data, sizeof(*data));
482
		sprintf(buf, "setup data %x", data->type);
483
		memblock_x86_reserve_range(pa_data, pa_data+sizeof(*data)+data->len, buf);
484 485 486 487 488
		pa_data = data->next;
		early_iounmap(data, sizeof(*data));
	}
}

489 490 491 492 493
/*
 * --------- Crashkernel reservation ------------------------------
 */

#ifdef CONFIG_KEXEC
494

495 496 497 498
static inline unsigned long long get_total_mem(void)
{
	unsigned long long total;

499
	total = max_pfn - min_low_pfn;
500 501 502 503

	return total << PAGE_SHIFT;
}

504 505 506 507 508 509 510 511 512 513 514 515
/*
 * 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

516
static void __init reserve_crashkernel(void)
517 518 519 520 521 522 523 524 525
{
	unsigned long long total_mem;
	unsigned long long crash_size, crash_base;
	int ret;

	total_mem = get_total_mem();

	ret = parse_crashkernel(boot_command_line, total_mem,
			&crash_size, &crash_base);
526 527 528 529 530
	if (ret != 0 || crash_size <= 0)
		return;

	/* 0 means: find the address automatically */
	if (crash_base <= 0) {
531 532
		const unsigned long long alignment = 16<<20;	/* 16M */

533
		/*
534
		 *  kexec want bzImage is below CRASH_KERNEL_ADDR_MAX
535 536
		 */
		crash_base = memblock_find_in_range(alignment,
537
			       CRASH_KERNEL_ADDR_MAX, crash_size, alignment);
538

539
		if (crash_base == MEMBLOCK_ERROR) {
540
			pr_info("crashkernel reservation failed - No suitable area found.\n");
541 542
			return;
		}
543
	} else {
544 545
		unsigned long long start;

546 547
		start = memblock_find_in_range(crash_base,
				 crash_base + crash_size, crash_size, 1<<20);
548 549
		if (start != crash_base) {
			pr_info("crashkernel reservation failed - memory is in use.\n");
550 551
			return;
		}
552
	}
553
	memblock_x86_reserve_range(crash_base, crash_base + crash_size, "CRASH KERNEL");
554

555 556 557 558 559
	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));
560

561 562 563
	crashk_res.start = crash_base;
	crashk_res.end   = crash_base + crash_size - 1;
	insert_resource(&iomem_resource, &crashk_res);
564 565
}
#else
566
static void __init reserve_crashkernel(void)
567 568 569 570
{
}
#endif

571 572 573 574 575 576 577 578 579 580 581 582 583 584 585 586 587 588 589 590 591 592 593
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 }
};

594
void __init reserve_standard_io_resources(void)
595 596 597 598 599 600 601 602 603
{
	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]);

}

604 605 606 607 608 609 610
/*
 * Note: elfcorehdr_addr is not just limited to vmcore. It is also used by
 * is_kdump_kernel() to determine if we are booting after a panic. Hence
 * ifdef it under CONFIG_CRASH_DUMP and not CONFIG_PROC_VMCORE.
 */

#ifdef CONFIG_CRASH_DUMP
611 612 613 614 615 616 617 618 619 620 621 622 623 624 625
/* elfcorehdr= specifies the location of elf core header
 * stored by the crashed kernel. This option will be passed
 * by kexec loader to the capture kernel.
 */
static int __init setup_elfcorehdr(char *arg)
{
	char *end;
	if (!arg)
		return -EINVAL;
	elfcorehdr_addr = memparse(arg, &end);
	return end > arg ? 0 : -EINVAL;
}
early_param("elfcorehdr", setup_elfcorehdr);
#endif

626 627 628 629 630 631 632
static __init void reserve_ibft_region(void)
{
	unsigned long addr, size = 0;

	addr = find_ibft_region(&size);

	if (size)
633
		memblock_x86_reserve_range(addr, addr + size, "* ibft");
634 635
}

636
static unsigned reserve_low = CONFIG_X86_RESERVE_LOW << 10;
637

638 639 640 641 642 643
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.
644 645 646
	 *
	 * This typically reserves additional memory (64KiB by default)
	 * since some BIOSes are known to corrupt low memory.  See the
647
	 * Kconfig help text for X86_RESERVE_LOW.
648
	 */
649
	e820_update_range(0, ALIGN(reserve_low, PAGE_SIZE),
650 651
			  E820_RAM, E820_RESERVED);

652 653 654 655 656 657 658 659 660
	/*
	 * 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);
}

661 662 663 664 665 666 667 668 669 670 671 672 673 674 675 676 677 678 679 680 681 682
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);

683 684 685 686 687 688 689 690 691
static u64 __init get_max_mapped(void)
{
	u64 end = max_pfn_mapped;

	end <<= PAGE_SHIFT;

	return end;
}

L
Linus Torvalds 已提交
692 693 694 695 696 697 698
/*
 * 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.
 */
699 700 701 702 703 704
/*
 * 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
705 706
void __init setup_arch(char **cmdline_p)
{
D
David Rientjes 已提交
707
	int acpi = 0;
708
	int amd = 0;
709
	unsigned long flags;
D
David Rientjes 已提交
710

711
#ifdef CONFIG_X86_32
L
Linus Torvalds 已提交
712
	memcpy(&boot_cpu_data, &new_cpu_data, sizeof(new_cpu_data));
I
Ingo Molnar 已提交
713
	visws_early_detect();
714 715 716 717 718 719 720 721 722 723 724

	/*
	 * 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();
725 726 727
#else
	printk(KERN_INFO "Command line: %s\n", boot_command_line);
#endif
L
Linus Torvalds 已提交
728

729 730 731 732
	/*
	 * If we have OLPC OFW, we might end up relocating the fixmap due to
	 * reserve_top(), so do this before touching the ioremap area.
	 */
733 734
	olpc_ofw_detect();

735
	early_trap_init();
736
	early_cpu_init();
737 738
	early_ioremap_init();

739 740
	setup_olpc_ofw_pgd();

741 742 743
	ROOT_DEV = old_decode_dev(boot_params.hdr.root_dev);
	screen_info = boot_params.screen_info;
	edid_info = boot_params.edid_info;
744
#ifdef CONFIG_X86_32
745 746
	apm_info.bios = boot_params.apm_bios_info;
	ist_info = boot_params.ist_info;
747
	if (boot_params.sys_desc_table.length != 0) {
748 749 750 751
		set_mca_bus(boot_params.sys_desc_table.table[3] & 0x2);
		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
752
	}
753 754
#endif
	saved_video_mode = boot_params.hdr.vid_mode;
755
	bootloader_type = boot_params.hdr.type_of_loader;
756 757 758 759 760 761
	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
762 763

#ifdef CONFIG_BLK_DEV_RAM
764 765 766
	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
767
#endif
768 769
#ifdef CONFIG_EFI
	if (!strncmp((char *)&boot_params.efi_info.efi_loader_signature,
770 771 772 773 774 775
#ifdef CONFIG_X86_32
		     "EL32",
#else
		     "EL64",
#endif
	 4)) {
776
		efi_enabled = 1;
777
		efi_memblock_x86_reserve_range();
778 779 780
	}
#endif

781
	x86_init.oem.arch_setup();
782

783
	iomem_resource.end = (1ULL << boot_cpu_data.x86_phys_bits) - 1;
784
	setup_memory_map();
785
	parse_setup_data();
786 787
	/* update the e820_saved too */
	e820_reserve_setup_data();
788

L
Linus Torvalds 已提交
789 790
	copy_edd();

791
	if (!boot_params.hdr.root_flags)
L
Linus Torvalds 已提交
792 793 794 795
		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;
796
	init_mm.brk = _brk_end;
L
Linus Torvalds 已提交
797 798 799 800 801

	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
802 803
	bss_resource.start = virt_to_phys(&__bss_start);
	bss_resource.end = virt_to_phys(&__bss_stop)-1;
L
Linus Torvalds 已提交
804

805 806 807 808 809 810 811 812 813 814 815 816 817
#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

818 819 820 821
	strlcpy(command_line, boot_command_line, COMMAND_LINE_SIZE);
	*cmdline_p = command_line;

	/*
822 823 824 825 826
	 * 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.
827
	 */
828
	x86_configure_nx();
829 830 831

	parse_early_param();

832
	x86_report_nx();
833

834
	/* after early param, so could get panic from serial */
835
	memblock_x86_reserve_range_setup_data();
836

837
	if (acpi_mps_check()) {
I
Ingo Molnar 已提交
838
#ifdef CONFIG_X86_LOCAL_APIC
839
		disable_apic = 1;
I
Ingo Molnar 已提交
840
#endif
841
		setup_clear_cpu_cap(X86_FEATURE_APIC);
842 843
	}

844 845 846 847 848
#ifdef CONFIG_PCI
	if (pci_early_dump_regs)
		early_dump_pci_devices();
#endif

849
	finish_e820_parsing();
850

B
Brian Maly 已提交
851 852 853
	if (efi_enabled)
		efi_init();

854 855
	dmi_scan_machine();

856 857 858 859
	/*
	 * VMware detection requires dmi to be available, so this
	 * needs to be done after dmi_scan_machine, for the BP.
	 */
860
	init_hypervisor_platform();
861

862
	x86_init.resources.probe_roms();
863 864 865 866 867 868

	/* 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);

869
	trim_bios_range();
870
#ifdef CONFIG_X86_32
Y
Yinghai Lu 已提交
871 872 873 874 875 876 877
	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");
	}
878 879 880
#else
	early_gart_iommu_check();
#endif
Y
Yinghai Lu 已提交
881

882 883 884 885
	/*
	 * partially used pages are not usable - thus
	 * we are rounding upwards:
	 */
886
	max_pfn = e820_end_of_ram_pfn();
887

888 889
	/* update e820 for memory not covered by WB MTRRs */
	mtrr_bp_init();
890
	if (mtrr_trim_uncached_memory(max_pfn))
891
		max_pfn = e820_end_of_ram_pfn();
892

893
#ifdef CONFIG_X86_32
894
	/* max_low_pfn get updated here */
895
	find_low_pfn_range();
896 897 898
#else
	num_physpages = max_pfn;

Y
Yinghai Lu 已提交
899
	check_x2apic();
900 901 902

	/* How many end-of-memory variables you have, grandma! */
	/* need this before calling reserve_initrd */
903 904 905 906 907
	if (max_pfn > (1UL<<(32 - PAGE_SHIFT)))
		max_low_pfn = e820_end_of_low_ram_pfn();
	else
		max_low_pfn = max_pfn;

908
	high_memory = (void *)__va(max_pfn * PAGE_SIZE - 1) + 1;
909 910
#endif

911 912 913 914 915
	/*
	 * Find and reserve possible boot-time SMP configuration:
	 */
	find_smp_config();

916 917
	reserve_ibft_region();

918 919 920 921 922 923 924 925 926 927 928 929 930 931 932 933 934 935 936 937
	/*
	 * Need to conclude brk, before memblock_x86_fill()
	 *  it could use memblock_find_in_range, could overlap with
	 *  brk area.
	 */
	reserve_brk();

	memblock.current_limit = get_max_mapped();
	memblock_x86_fill();

	/* preallocate 4k for mptable mpc */
	early_reserve_e820_mpc_new();

#ifdef CONFIG_X86_CHECK_BIOS_CORRUPTION
	setup_bios_corruption_check();
#endif

	printk(KERN_DEBUG "initial memory mapped : 0 - %08lx\n",
			max_pfn_mapped<<PAGE_SHIFT);

938 939
	reserve_trampoline_memory();

940 941 942 943 944 945 946
#ifdef CONFIG_ACPI_SLEEP
	/*
	 * Reserve low memory region for sleep support.
	 * even before init_memory_mapping
	 */
	acpi_reserve_wakeup_memory();
#endif
947 948
	init_gbpages();

949
	/* max_pfn_mapped is updated here */
950 951 952 953 954 955 956 957 958 959 960
	max_low_pfn_mapped = init_memory_mapping(0, max_low_pfn<<PAGE_SHIFT);
	max_pfn_mapped = max_low_pfn_mapped;

#ifdef CONFIG_X86_64
	if (max_pfn > max_low_pfn) {
		max_pfn_mapped = init_memory_mapping(1UL<<32,
						     max_pfn<<PAGE_SHIFT);
		/* can we preseve max_low_pfn ?*/
		max_low_pfn = max_pfn;
	}
#endif
961
	memblock.current_limit = get_max_mapped();
962

963 964 965 966 967 968 969 970 971
	/*
	 * 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

972 973
	reserve_initrd();

974 975
	reserve_crashkernel();

976 977
	vsmp_init();

978 979 980 981 982 983 984
	io_delay_init();

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

985 986
	early_acpi_boot_init();

987
#ifdef CONFIG_ACPI_NUMA
Y
Yinghai Lu 已提交
988 989 990
	/*
	 * Parse SRAT to discover nodes.
	 */
991
	acpi = acpi_numa_init();
992 993
#endif

994
#ifdef CONFIG_AMD_NUMA
995
	if (!acpi)
996
		amd = !amd_numa_init(0, max_pfn);
D
David Rientjes 已提交
997 998
#endif

999
	initmem_init(0, max_pfn, acpi, amd);
1000
	memblock_find_dma_reserve();
1001 1002
	dma32_reserve_bootmem();

1003 1004 1005 1006
#ifdef CONFIG_KVM_CLOCK
	kvmclock_init();
#endif

1007
	x86_init.paging.pagetable_setup_start(swapper_pg_dir);
L
Linus Torvalds 已提交
1008
	paging_init();
1009
	x86_init.paging.pagetable_setup_done(swapper_pg_dir);
1010

1011 1012 1013 1014 1015 1016
#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
1017

1018 1019
	tboot_probe();

1020 1021 1022 1023
#ifdef CONFIG_X86_64
	map_vsyscall();
#endif

1024
	generic_apic_probe();
L
Linus Torvalds 已提交
1025

1026
	early_quirks();
1027

1028 1029 1030
	/*
	 * Read APIC and some other early information from ACPI tables.
	 */
L
Linus Torvalds 已提交
1031
	acpi_boot_init();
1032

F
Feng Tang 已提交
1033
	sfi_init();
1034

1035 1036 1037
	/*
	 * get boot-time SMP configuration:
	 */
1038 1039
	if (smp_found_config)
		get_smp_config();
1040

1041
	prefill_possible_map();
Y
Yinghai Lu 已提交
1042

1043 1044 1045 1046
#ifdef CONFIG_X86_64
	init_cpu_to_node();
#endif

1047
	init_apic_mappings();
1048
	ioapic_and_gsi_init();
1049

1050
	kvm_guest_init();
L
Linus Torvalds 已提交
1051

1052
	e820_reserve_resources();
1053
	e820_mark_nosave_regions(max_low_pfn);
L
Linus Torvalds 已提交
1054

1055
	x86_init.resources.reserve_resources();
1056 1057 1058

	e820_setup_gap();

L
Linus Torvalds 已提交
1059 1060 1061 1062 1063 1064 1065 1066
#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
1067
	x86_init.oem.banner();
1068 1069

	mcheck_init();
1070 1071 1072 1073

	local_irq_save(flags);
	arch_init_ideal_nop5();
	local_irq_restore(flags);
L
Linus Torvalds 已提交
1074
}
1075

1076 1077
#ifdef CONFIG_X86_32

1078 1079 1080 1081 1082
static struct resource video_ram_resource = {
	.name	= "Video RAM area",
	.start	= 0xa0000,
	.end	= 0xbffff,
	.flags	= IORESOURCE_BUSY | IORESOURCE_MEM
1083 1084
};

1085
void __init i386_reserve_resources(void)
1086
{
1087 1088
	request_resource(&iomem_resource, &video_ram_resource);
	reserve_standard_io_resources();
1089 1090 1091
}

#endif /* CONFIG_X86_32 */