reboot.c 21.7 KB
Newer Older
L
Linus Torvalds 已提交
1
#include <linux/module.h>
2
#include <linux/reboot.h>
M
Miguel Boton 已提交
3 4 5
#include <linux/init.h>
#include <linux/pm.h>
#include <linux/efi.h>
6
#include <linux/dmi.h>
7
#include <linux/sched.h>
8
#include <linux/tboot.h>
M
Miguel Boton 已提交
9 10
#include <acpi/reboot.h>
#include <asm/io.h>
L
Linus Torvalds 已提交
11
#include <asm/apic.h>
12
#include <asm/desc.h>
M
Miguel Boton 已提交
13
#include <asm/hpet.h>
J
Jeremy Fitzhardinge 已提交
14
#include <asm/pgtable.h>
15
#include <asm/proto.h>
16
#include <asm/reboot_fixups.h>
17
#include <asm/reboot.h>
18
#include <asm/pci_x86.h>
19
#include <asm/virtext.h>
20
#include <asm/cpu.h>
L
Linus Torvalds 已提交
21

M
Miguel Boton 已提交
22 23 24 25
#ifdef CONFIG_X86_32
# include <linux/ctype.h>
# include <linux/mc146818rtc.h>
#else
26
# include <asm/x86_init.h>
M
Miguel Boton 已提交
27 28
#endif

L
Linus Torvalds 已提交
29 30 31 32
/*
 * Power off function, if any
 */
void (*pm_power_off)(void);
33
EXPORT_SYMBOL(pm_power_off);
L
Linus Torvalds 已提交
34

J
Jan Beulich 已提交
35
static const struct desc_ptr no_idt = {};
L
Linus Torvalds 已提交
36
static int reboot_mode;
37
enum reboot_type reboot_type = BOOT_KBD;
M
Miguel Boton 已提交
38
int reboot_force;
L
Linus Torvalds 已提交
39

M
Miguel Boton 已提交
40
#if defined(CONFIG_X86_32) && defined(CONFIG_SMP)
L
Linus Torvalds 已提交
41 42
static int reboot_cpu = -1;
#endif
M
Miguel Boton 已提交
43

44 45 46 47 48 49
/* This is set if we need to go through the 'emergency' path.
 * When machine_emergency_restart() is called, we may be on
 * an inconsistent state and won't be able to do a clean cleanup
 */
static int reboot_emergency;

50 51 52 53
/* This is set by the PCI code if either type 1 or type 2 PCI is detected */
bool port_cf9_safe = false;

/* reboot=b[ios] | s[mp] | t[riple] | k[bd] | e[fi] [, [w]arm | [c]old] | p[ci]
M
Miguel Boton 已提交
54 55 56 57 58 59 60 61
   warm   Don't set the cold reboot flag
   cold   Set the cold reboot flag
   bios   Reboot by jumping through the BIOS (only for X86_32)
   smp    Reboot by executing reset on BSP or other CPU (only for X86_32)
   triple Force a triple fault (init)
   kbd    Use the keyboard controller. cold reset (default)
   acpi   Use the RESET_REG in the FADT
   efi    Use efi reset_system runtime service
62
   pci    Use the so-called "PCI reset register", CF9
M
Miguel Boton 已提交
63 64
   force  Avoid anything that could hang.
 */
L
Linus Torvalds 已提交
65 66
static int __init reboot_setup(char *str)
{
M
Miguel Boton 已提交
67
	for (;;) {
L
Linus Torvalds 已提交
68
		switch (*str) {
M
Miguel Boton 已提交
69
		case 'w':
L
Linus Torvalds 已提交
70 71
			reboot_mode = 0x1234;
			break;
M
Miguel Boton 已提交
72 73 74

		case 'c':
			reboot_mode = 0;
L
Linus Torvalds 已提交
75
			break;
M
Miguel Boton 已提交
76 77

#ifdef CONFIG_X86_32
L
Linus Torvalds 已提交
78
#ifdef CONFIG_SMP
M
Miguel Boton 已提交
79
		case 's':
80
			if (isdigit(*(str+1))) {
L
Linus Torvalds 已提交
81
				reboot_cpu = (int) (*(str+1) - '0');
82
				if (isdigit(*(str+2)))
L
Linus Torvalds 已提交
83 84
					reboot_cpu = reboot_cpu*10 + (int)(*(str+2) - '0');
			}
M
Miguel Boton 已提交
85 86
				/* we will leave sorting out the final value
				   when we are ready to reboot, since we might not
87
				   have detected BSP APIC ID or smp_num_cpu */
L
Linus Torvalds 已提交
88
			break;
M
Miguel Boton 已提交
89 90 91
#endif /* CONFIG_SMP */

		case 'b':
L
Linus Torvalds 已提交
92
#endif
M
Miguel Boton 已提交
93 94 95 96
		case 'a':
		case 'k':
		case 't':
		case 'e':
97
		case 'p':
M
Miguel Boton 已提交
98 99 100 101 102 103
			reboot_type = *str;
			break;

		case 'f':
			reboot_force = 1;
			break;
L
Linus Torvalds 已提交
104
		}
M
Miguel Boton 已提交
105 106 107

		str = strchr(str, ',');
		if (str)
L
Linus Torvalds 已提交
108 109 110 111 112 113 114 115 116
			str++;
		else
			break;
	}
	return 1;
}

__setup("reboot=", reboot_setup);

M
Miguel Boton 已提交
117 118

#ifdef CONFIG_X86_32
L
Linus Torvalds 已提交
119 120 121 122 123 124
/*
 * Reboot options and system auto-detection code provided by
 * Dell Inc. so their systems "just work". :-)
 */

/*
M
Miguel Boton 已提交
125 126
 * Some machines require the "reboot=b"  commandline option,
 * this quirk makes that automatic.
L
Linus Torvalds 已提交
127
 */
128
static int __init set_bios_reboot(const struct dmi_system_id *d)
L
Linus Torvalds 已提交
129
{
M
Miguel Boton 已提交
130 131
	if (reboot_type != BOOT_BIOS) {
		reboot_type = BOOT_BIOS;
L
Linus Torvalds 已提交
132 133 134 135 136 137
		printk(KERN_INFO "%s series board detected. Selecting BIOS-method for reboots.\n", d->ident);
	}
	return 0;
}

static struct dmi_system_id __initdata reboot_dmi_table[] = {
138 139 140 141 142 143 144 145
	{	/* Handle problems with rebooting on Dell E520's */
		.callback = set_bios_reboot,
		.ident = "Dell E520",
		.matches = {
			DMI_MATCH(DMI_SYS_VENDOR, "Dell Inc."),
			DMI_MATCH(DMI_PRODUCT_NAME, "Dell DM061"),
		},
	},
L
Linus Torvalds 已提交
146
	{	/* Handle problems with rebooting on Dell 1300's */
147
		.callback = set_bios_reboot,
L
Linus Torvalds 已提交
148 149 150 151 152 153 154 155 156 157 158 159 160 161
		.ident = "Dell PowerEdge 1300",
		.matches = {
			DMI_MATCH(DMI_SYS_VENDOR, "Dell Computer Corporation"),
			DMI_MATCH(DMI_PRODUCT_NAME, "PowerEdge 1300/"),
		},
	},
	{	/* Handle problems with rebooting on Dell 300's */
		.callback = set_bios_reboot,
		.ident = "Dell PowerEdge 300",
		.matches = {
			DMI_MATCH(DMI_SYS_VENDOR, "Dell Computer Corporation"),
			DMI_MATCH(DMI_PRODUCT_NAME, "PowerEdge 300/"),
		},
	},
162 163 164 165 166 167 168 169
	{       /* Handle problems with rebooting on Dell Optiplex 745's SFF*/
		.callback = set_bios_reboot,
		.ident = "Dell OptiPlex 745",
		.matches = {
			DMI_MATCH(DMI_SYS_VENDOR, "Dell Inc."),
			DMI_MATCH(DMI_PRODUCT_NAME, "OptiPlex 745"),
		},
	},
170 171 172 173 174 175 176 177 178
	{       /* Handle problems with rebooting on Dell Optiplex 745's DFF*/
		.callback = set_bios_reboot,
		.ident = "Dell OptiPlex 745",
		.matches = {
			DMI_MATCH(DMI_SYS_VENDOR, "Dell Inc."),
			DMI_MATCH(DMI_PRODUCT_NAME, "OptiPlex 745"),
			DMI_MATCH(DMI_BOARD_NAME, "0MM599"),
		},
	},
179 180 181 182 183 184 185 186 187
	{       /* Handle problems with rebooting on Dell Optiplex 745 with 0KW626 */
		.callback = set_bios_reboot,
		.ident = "Dell OptiPlex 745",
		.matches = {
			DMI_MATCH(DMI_SYS_VENDOR, "Dell Inc."),
			DMI_MATCH(DMI_PRODUCT_NAME, "OptiPlex 745"),
			DMI_MATCH(DMI_BOARD_NAME, "0KW626"),
		},
	},
188 189 190 191 192 193 194 195 196
	{   /* Handle problems with rebooting on Dell Optiplex 330 with 0KP561 */
		.callback = set_bios_reboot,
		.ident = "Dell OptiPlex 330",
		.matches = {
			DMI_MATCH(DMI_SYS_VENDOR, "Dell Inc."),
			DMI_MATCH(DMI_PRODUCT_NAME, "OptiPlex 330"),
			DMI_MATCH(DMI_BOARD_NAME, "0KP561"),
		},
	},
197 198 199 200 201 202 203 204 205
	{   /* Handle problems with rebooting on Dell Optiplex 360 with 0T656F */
		.callback = set_bios_reboot,
		.ident = "Dell OptiPlex 360",
		.matches = {
			DMI_MATCH(DMI_SYS_VENDOR, "Dell Inc."),
			DMI_MATCH(DMI_PRODUCT_NAME, "OptiPlex 360"),
			DMI_MATCH(DMI_BOARD_NAME, "0T656F"),
		},
	},
206 207 208 209 210 211 212 213 214
	{	/* Handle problems with rebooting on Dell OptiPlex 760 with 0G919G*/
		.callback = set_bios_reboot,
		.ident = "Dell OptiPlex 760",
		.matches = {
			DMI_MATCH(DMI_SYS_VENDOR, "Dell Inc."),
			DMI_MATCH(DMI_PRODUCT_NAME, "OptiPlex 760"),
			DMI_MATCH(DMI_BOARD_NAME, "0G919G"),
		},
	},
L
Linus Torvalds 已提交
215 216 217 218 219 220 221 222
	{	/* Handle problems with rebooting on Dell 2400's */
		.callback = set_bios_reboot,
		.ident = "Dell PowerEdge 2400",
		.matches = {
			DMI_MATCH(DMI_SYS_VENDOR, "Dell Computer Corporation"),
			DMI_MATCH(DMI_PRODUCT_NAME, "PowerEdge 2400"),
		},
	},
223 224 225 226 227 228 229 230
	{	/* Handle problems with rebooting on Dell T5400's */
		.callback = set_bios_reboot,
		.ident = "Dell Precision T5400",
		.matches = {
			DMI_MATCH(DMI_SYS_VENDOR, "Dell Inc."),
			DMI_MATCH(DMI_PRODUCT_NAME, "Precision WorkStation T5400"),
		},
	},
231 232 233 234 235 236 237 238
	{	/* Handle problems with rebooting on Dell T7400's */
		.callback = set_bios_reboot,
		.ident = "Dell Precision T7400",
		.matches = {
			DMI_MATCH(DMI_SYS_VENDOR, "Dell Inc."),
			DMI_MATCH(DMI_PRODUCT_NAME, "Precision WorkStation T7400"),
		},
	},
239
	{	/* Handle problems with rebooting on HP laptops */
240
		.callback = set_bios_reboot,
241
		.ident = "HP Compaq Laptop",
242 243
		.matches = {
			DMI_MATCH(DMI_SYS_VENDOR, "Hewlett-Packard"),
244
			DMI_MATCH(DMI_PRODUCT_NAME, "HP Compaq"),
245 246
		},
	},
247 248 249 250 251 252 253 254
	{	/* Handle problems with rebooting on Dell XPS710 */
		.callback = set_bios_reboot,
		.ident = "Dell XPS710",
		.matches = {
			DMI_MATCH(DMI_SYS_VENDOR, "Dell Inc."),
			DMI_MATCH(DMI_PRODUCT_NAME, "Dell XPS710"),
		},
	},
255 256 257 258 259 260 261 262
	{	/* Handle problems with rebooting on Dell DXP061 */
		.callback = set_bios_reboot,
		.ident = "Dell DXP061",
		.matches = {
			DMI_MATCH(DMI_SYS_VENDOR, "Dell Inc."),
			DMI_MATCH(DMI_PRODUCT_NAME, "Dell DXP061"),
		},
	},
263 264 265 266 267 268 269 270
	{	/* Handle problems with rebooting on Sony VGN-Z540N */
		.callback = set_bios_reboot,
		.ident = "Sony VGN-Z540N",
		.matches = {
			DMI_MATCH(DMI_SYS_VENDOR, "Sony Corporation"),
			DMI_MATCH(DMI_PRODUCT_NAME, "VGN-Z540N"),
		},
	},
271 272 273 274 275 276 277 278
	{	/* Handle problems with rebooting on CompuLab SBC-FITPC2 */
		.callback = set_bios_reboot,
		.ident = "CompuLab SBC-FITPC2",
		.matches = {
			DMI_MATCH(DMI_SYS_VENDOR, "CompuLab"),
			DMI_MATCH(DMI_PRODUCT_NAME, "SBC-FITPC2"),
		},
	},
279 280 281 282 283 284 285 286
	{       /* Handle problems with rebooting on ASUS P4S800 */
		.callback = set_bios_reboot,
		.ident = "ASUS P4S800",
		.matches = {
			DMI_MATCH(DMI_BOARD_VENDOR, "ASUSTeK Computer INC."),
			DMI_MATCH(DMI_BOARD_NAME, "P4S800"),
		},
	},
L
Linus Torvalds 已提交
287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302
	{ }
};

static int __init reboot_init(void)
{
	dmi_check_system(reboot_dmi_table);
	return 0;
}
core_initcall(reboot_init);

/* The following code and data reboots the machine by switching to real
   mode and jumping to the BIOS reset entry point, as if the CPU has
   really been reset.  The previous version asked the keyboard
   controller to pulse the CPU reset line, which is more thorough, but
   doesn't work with at least one type of 486 motherboard.  It is easy
   to stop this code working; hence the copious comments. */
J
Jan Beulich 已提交
303
static const unsigned long long
L
Linus Torvalds 已提交
304 305 306
real_mode_gdt_entries [3] =
{
	0x0000000000000000ULL,	/* Null descriptor */
J
Jan Beulich 已提交
307 308
	0x00009b000000ffffULL,	/* 16-bit real-mode 64k code at 0x00000000 */
	0x000093000100ffffULL	/* 16-bit real-mode 64k data at 0x00000100 */
L
Linus Torvalds 已提交
309 310
};

J
Jan Beulich 已提交
311
static const struct desc_ptr
312
real_mode_gdt = { sizeof (real_mode_gdt_entries) - 1, (long)real_mode_gdt_entries },
M
Miguel Boton 已提交
313
real_mode_idt = { 0x3ff, 0 };
L
Linus Torvalds 已提交
314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332

/* This is 16-bit protected mode code to disable paging and the cache,
   switch to real mode and jump to the BIOS reset code.

   The instruction that switches to real mode by writing to CR0 must be
   followed immediately by a far jump instruction, which set CS to a
   valid value for real mode, and flushes the prefetch queue to avoid
   running instructions that have already been decoded in protected
   mode.

   Clears all the flags except ET, especially PG (paging), PE
   (protected-mode enable) and TS (task switch for coprocessor state
   save).  Flushes the TLB after paging has been disabled.  Sets CD and
   NW, to disable the cache on a 486, and invalidates the cache.  This
   is more like the state of a 486 after reset.  I don't know if
   something else should be done for other chips.

   More could be done here to set up the registers as if a CPU reset had
   occurred; hopefully real BIOSs don't assume much. */
J
Jan Beulich 已提交
333
static const unsigned char real_mode_switch [] =
L
Linus Torvalds 已提交
334 335 336 337 338 339 340 341 342 343 344 345 346
{
	0x66, 0x0f, 0x20, 0xc0,			/*    movl  %cr0,%eax        */
	0x66, 0x83, 0xe0, 0x11,			/*    andl  $0x00000011,%eax */
	0x66, 0x0d, 0x00, 0x00, 0x00, 0x60,	/*    orl   $0x60000000,%eax */
	0x66, 0x0f, 0x22, 0xc0,			/*    movl  %eax,%cr0        */
	0x66, 0x0f, 0x22, 0xd8,			/*    movl  %eax,%cr3        */
	0x66, 0x0f, 0x20, 0xc3,			/*    movl  %cr0,%ebx        */
	0x66, 0x81, 0xe3, 0x00, 0x00, 0x00, 0x60,	/*    andl  $0x60000000,%ebx */
	0x74, 0x02,				/*    jz    f                */
	0x0f, 0x09,				/*    wbinvd                 */
	0x24, 0x10,				/* f: andb  $0x10,al         */
	0x66, 0x0f, 0x22, 0xc0			/*    movl  %eax,%cr0        */
};
J
Jan Beulich 已提交
347
static const unsigned char jump_to_bios [] =
L
Linus Torvalds 已提交
348 349 350 351 352 353 354 355 356
{
	0xea, 0x00, 0x00, 0xff, 0xff		/*    ljmp  $0xffff,$0x0000  */
};

/*
 * Switch to real mode and then execute the code
 * specified by the code and length parameters.
 * We assume that length will aways be less that 100!
 */
J
Jan Beulich 已提交
357
void machine_real_restart(const unsigned char *code, int length)
L
Linus Torvalds 已提交
358 359 360 361 362 363 364 365 366 367 368 369
{
	local_irq_disable();

	/* Write zero to CMOS register number 0x0f, which the BIOS POST
	   routine will recognize as telling it to do a proper reboot.  (Well
	   that's what this book in front of me says -- it may only apply to
	   the Phoenix BIOS though, it's not clear).  At the same time,
	   disable NMIs by setting the top bit in the CMOS address register,
	   as we're about to do peculiar things to the CPU.  I'm not sure if
	   `outb_p' is needed instead of just `outb'.  Use it to be on the
	   safe side.  (Yes, CMOS_WRITE does outb_p's. -  Paul G.)
	 */
370
	spin_lock(&rtc_lock);
L
Linus Torvalds 已提交
371
	CMOS_WRITE(0x00, 0x8f);
372
	spin_unlock(&rtc_lock);
L
Linus Torvalds 已提交
373 374

	/*
375
	 * Switch back to the initial page table.
L
Linus Torvalds 已提交
376
	 */
377
	load_cr3(initial_page_table);
L
Linus Torvalds 已提交
378 379 380 381 382 383 384 385 386 387 388 389 390

	/* Write 0x1234 to absolute memory location 0x472.  The BIOS reads
	   this on booting to tell it to "Bypass memory test (also warm
	   boot)".  This seems like a fairly standard thing that gets set by
	   REBOOT.COM programs, and the previous reset routine did this
	   too. */
	*((unsigned short *)0x472) = reboot_mode;

	/* For the switch to real mode, copy some code to low memory.  It has
	   to be in the first 64k because it is running in 16-bit mode, and it
	   has to have the same physical and virtual address, because it turns
	   off paging.  Copy it near the end of the first page, out of the way
	   of BIOS variables. */
M
Miguel Boton 已提交
391
	memcpy((void *)(0x1000 - sizeof(real_mode_switch) - 100),
L
Linus Torvalds 已提交
392
		real_mode_switch, sizeof (real_mode_switch));
M
Miguel Boton 已提交
393
	memcpy((void *)(0x1000 - 100), code, length);
L
Linus Torvalds 已提交
394 395

	/* Set up the IDT for real mode. */
396
	load_idt(&real_mode_idt);
L
Linus Torvalds 已提交
397 398 399 400

	/* Set up a GDT from which we can load segment descriptors for real
	   mode.  The GDT is not used in real mode; it is just needed here to
	   prepare the descriptors. */
401
	load_gdt(&real_mode_gdt);
L
Linus Torvalds 已提交
402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419

	/* Load the data segment registers, and thus the descriptors ready for
	   real mode.  The base address of each segment is 0x100, 16 times the
	   selector value being loaded here.  This is so that the segment
	   registers don't have to be reloaded after switching to real mode:
	   the values are consistent for real mode operation already. */
	__asm__ __volatile__ ("movl $0x0010,%%eax\n"
				"\tmovl %%eax,%%ds\n"
				"\tmovl %%eax,%%es\n"
				"\tmovl %%eax,%%fs\n"
				"\tmovl %%eax,%%gs\n"
				"\tmovl %%eax,%%ss" : : : "eax");

	/* Jump to the 16-bit code that we copied earlier.  It disables paging
	   and the cache, switches to real mode, and jumps to the BIOS reset
	   entry point. */
	__asm__ __volatile__ ("ljmp $0x0008,%0"
				:
M
Miguel Boton 已提交
420
				: "i" ((void *)(0x1000 - sizeof (real_mode_switch) - 100)));
L
Linus Torvalds 已提交
421
}
422 423 424
#ifdef CONFIG_APM_MODULE
EXPORT_SYMBOL(machine_real_restart);
#endif
L
Linus Torvalds 已提交
425

M
Miguel Boton 已提交
426 427
#endif /* CONFIG_X86_32 */

428
/*
429
 * Some Apple MacBook and MacBookPro's needs reboot=p to be able to reboot
430 431 432 433 434 435 436 437 438 439 440 441
 */
static int __init set_pci_reboot(const struct dmi_system_id *d)
{
	if (reboot_type != BOOT_CF9) {
		reboot_type = BOOT_CF9;
		printk(KERN_INFO "%s series board detected. "
		       "Selecting PCI-method for reboots.\n", d->ident);
	}
	return 0;
}

static struct dmi_system_id __initdata pci_reboot_dmi_table[] = {
442
	{	/* Handle problems with rebooting on Apple MacBook5 */
443
		.callback = set_pci_reboot,
444
		.ident = "Apple MacBook5",
445 446
		.matches = {
			DMI_MATCH(DMI_SYS_VENDOR, "Apple Inc."),
447
			DMI_MATCH(DMI_PRODUCT_NAME, "MacBook5"),
448 449
		},
	},
450
	{	/* Handle problems with rebooting on Apple MacBookPro5 */
451
		.callback = set_pci_reboot,
452
		.ident = "Apple MacBookPro5",
453 454
		.matches = {
			DMI_MATCH(DMI_SYS_VENDOR, "Apple Inc."),
455
			DMI_MATCH(DMI_PRODUCT_NAME, "MacBookPro5"),
456 457
		},
	},
458 459 460 461 462 463 464 465
	{	/* Handle problems with rebooting on Apple Macmini3,1 */
		.callback = set_pci_reboot,
		.ident = "Apple Macmini3,1",
		.matches = {
			DMI_MATCH(DMI_SYS_VENDOR, "Apple Inc."),
			DMI_MATCH(DMI_PRODUCT_NAME, "Macmini3,1"),
		},
	},
466 467 468 469 470 471 472 473
	{	/* Handle problems with rebooting on the iMac9,1. */
		.callback = set_pci_reboot,
		.ident = "Apple iMac9,1",
		.matches = {
			DMI_MATCH(DMI_SYS_VENDOR, "Apple Inc."),
			DMI_MATCH(DMI_PRODUCT_NAME, "iMac9,1"),
		},
	},
474 475 476 477 478 479 480 481 482 483
	{ }
};

static int __init pci_reboot_init(void)
{
	dmi_check_system(pci_reboot_dmi_table);
	return 0;
}
core_initcall(pci_reboot_init);

M
Miguel Boton 已提交
484 485 486 487
static inline void kb_wait(void)
{
	int i;

A
Alan Cox 已提交
488 489
	for (i = 0; i < 0x10000; i++) {
		if ((inb(0x64) & 0x02) == 0)
M
Miguel Boton 已提交
490
			break;
A
Alan Cox 已提交
491 492
		udelay(2);
	}
M
Miguel Boton 已提交
493 494
}

495 496 497 498 499 500 501 502 503 504 505 506 507 508 509 510 511 512 513 514 515 516 517 518 519 520 521 522 523 524 525 526 527 528 529 530 531 532 533 534 535 536
static void vmxoff_nmi(int cpu, struct die_args *args)
{
	cpu_emergency_vmxoff();
}

/* Use NMIs as IPIs to tell all CPUs to disable virtualization
 */
static void emergency_vmx_disable_all(void)
{
	/* Just make sure we won't change CPUs while doing this */
	local_irq_disable();

	/* We need to disable VMX on all CPUs before rebooting, otherwise
	 * we risk hanging up the machine, because the CPU ignore INIT
	 * signals when VMX is enabled.
	 *
	 * We can't take any locks and we may be on an inconsistent
	 * state, so we use NMIs as IPIs to tell the other CPUs to disable
	 * VMX and halt.
	 *
	 * For safety, we will avoid running the nmi_shootdown_cpus()
	 * stuff unnecessarily, but we don't have a way to check
	 * if other CPUs have VMX enabled. So we will call it only if the
	 * CPU we are running on has VMX enabled.
	 *
	 * We will miss cases where VMX is not enabled on all CPUs. This
	 * shouldn't do much harm because KVM always enable VMX on all
	 * CPUs anyway. But we can miss it on the small window where KVM
	 * is still enabling VMX.
	 */
	if (cpu_has_vmx() && cpu_vmx_enabled()) {
		/* Disable VMX on this CPU.
		 */
		cpu_vmxoff();

		/* Halt and disable VMX on the other CPUs */
		nmi_shootdown_cpus(vmxoff_nmi);

	}
}


I
Ingo Molnar 已提交
537 538 539 540
void __attribute__((weak)) mach_reboot_fixups(void)
{
}

541
static void native_machine_emergency_restart(void)
L
Linus Torvalds 已提交
542
{
M
Miguel Boton 已提交
543 544
	int i;

545 546 547
	if (reboot_emergency)
		emergency_vmx_disable_all();

548 549
	tboot_shutdown(TB_SHUTDOWN_REBOOT);

M
Miguel Boton 已提交
550 551 552 553 554 555 556
	/* Tell the BIOS if we want cold or warm reboot */
	*((unsigned short *)__va(0x472)) = reboot_mode;

	for (;;) {
		/* Could also try the reset bit in the Hammer NB */
		switch (reboot_type) {
		case BOOT_KBD:
I
Ingo Molnar 已提交
557 558
			mach_reboot_fixups(); /* for board specific fixups */

M
Miguel Boton 已提交
559 560 561 562 563 564 565 566
			for (i = 0; i < 10; i++) {
				kb_wait();
				udelay(50);
				outb(0xfe, 0x64); /* pulse reset low */
				udelay(50);
			}

		case BOOT_TRIPLE:
J
Jan Beulich 已提交
567
			load_idt(&no_idt);
M
Miguel Boton 已提交
568 569 570 571 572 573 574 575 576 577 578 579 580 581 582 583 584 585 586 587
			__asm__ __volatile__("int3");

			reboot_type = BOOT_KBD;
			break;

#ifdef CONFIG_X86_32
		case BOOT_BIOS:
			machine_real_restart(jump_to_bios, sizeof(jump_to_bios));

			reboot_type = BOOT_KBD;
			break;
#endif

		case BOOT_ACPI:
			acpi_reboot();
			reboot_type = BOOT_KBD;
			break;

		case BOOT_EFI:
			if (efi_enabled)
588 589 590
				efi.reset_system(reboot_mode ?
						 EFI_RESET_WARM :
						 EFI_RESET_COLD,
M
Miguel Boton 已提交
591
						 EFI_SUCCESS, 0, NULL);
592
			reboot_type = BOOT_KBD;
593
			break;
M
Miguel Boton 已提交
594

595 596 597
		case BOOT_CF9:
			port_cf9_safe = true;
			/* fall through */
M
Miguel Boton 已提交
598

599 600 601 602 603 604 605 606
		case BOOT_CF9_COND:
			if (port_cf9_safe) {
				u8 cf9 = inb(0xcf9) & ~6;
				outb(cf9|2, 0xcf9); /* Request hard reset */
				udelay(50);
				outb(cf9|6, 0xcf9); /* Actually do the reset */
				udelay(50);
			}
M
Miguel Boton 已提交
607 608 609 610 611 612
			reboot_type = BOOT_KBD;
			break;
		}
	}
}

613
void native_machine_shutdown(void)
M
Miguel Boton 已提交
614 615
{
	/* Stop the cpus and apics */
L
Linus Torvalds 已提交
616
#ifdef CONFIG_SMP
617 618

	/* The boot cpu is always logical cpu 0 */
619
	int reboot_cpu_id = 0;
620

M
Miguel Boton 已提交
621
#ifdef CONFIG_X86_32
622
	/* See if there has been given a command line override */
623
	if ((reboot_cpu != -1) && (reboot_cpu < nr_cpu_ids) &&
624
		cpu_online(reboot_cpu))
625
		reboot_cpu_id = reboot_cpu;
M
Miguel Boton 已提交
626
#endif
L
Linus Torvalds 已提交
627

M
Miguel Boton 已提交
628
	/* Make certain the cpu I'm about to reboot on is online */
629
	if (!cpu_online(reboot_cpu_id))
630 631 632
		reboot_cpu_id = smp_processor_id();

	/* Make certain I only run on the appropriate processor */
633
	set_cpus_allowed_ptr(current, cpumask_of(reboot_cpu_id));
634

M
Miguel Boton 已提交
635 636
	/* O.K Now that I'm on the appropriate processor,
	 * stop all of the others.
L
Linus Torvalds 已提交
637
	 */
638
	stop_other_cpus();
M
Miguel Boton 已提交
639
#endif
L
Linus Torvalds 已提交
640 641 642 643 644 645

	lapic_shutdown();

#ifdef CONFIG_X86_IO_APIC
	disable_IO_APIC();
#endif
M
Miguel Boton 已提交
646

O
OGAWA Hirofumi 已提交
647 648 649
#ifdef CONFIG_HPET_TIMER
	hpet_disable();
#endif
650

M
Miguel Boton 已提交
651
#ifdef CONFIG_X86_64
652
	x86_platform.iommu_shutdown();
M
Miguel Boton 已提交
653
#endif
654 655
}

656 657 658 659 660 661
static void __machine_emergency_restart(int emergency)
{
	reboot_emergency = emergency;
	machine_ops.emergency_restart();
}

662
static void native_machine_restart(char *__unused)
663
{
M
Miguel Boton 已提交
664
	printk("machine restart\n");
L
Linus Torvalds 已提交
665

M
Miguel Boton 已提交
666 667
	if (!reboot_force)
		machine_shutdown();
668
	__machine_emergency_restart(0);
669 670
}

671
static void native_machine_halt(void)
L
Linus Torvalds 已提交
672
{
673 674 675
	/* stop other cpus and apics */
	machine_shutdown();

676 677
	tboot_shutdown(TB_SHUTDOWN_HALT);

678 679
	/* stop this cpu */
	stop_this_cpu(NULL);
L
Linus Torvalds 已提交
680 681
}

682
static void native_machine_power_off(void)
L
Linus Torvalds 已提交
683
{
684
	if (pm_power_off) {
M
Miguel Boton 已提交
685 686
		if (!reboot_force)
			machine_shutdown();
L
Linus Torvalds 已提交
687
		pm_power_off();
688
	}
689 690
	/* a fallback in case there is no PM info available */
	tboot_shutdown(TB_SHUTDOWN_HALT);
L
Linus Torvalds 已提交
691 692
}

693
struct machine_ops machine_ops = {
694 695 696 697
	.power_off = native_machine_power_off,
	.shutdown = native_machine_shutdown,
	.emergency_restart = native_machine_emergency_restart,
	.restart = native_machine_restart,
698 699 700 701
	.halt = native_machine_halt,
#ifdef CONFIG_KEXEC
	.crash_shutdown = native_machine_crash_shutdown,
#endif
702
};
703 704 705 706 707 708 709 710 711 712 713 714 715

void machine_power_off(void)
{
	machine_ops.power_off();
}

void machine_shutdown(void)
{
	machine_ops.shutdown();
}

void machine_emergency_restart(void)
{
716
	__machine_emergency_restart(1);
717 718 719 720 721 722 723 724 725 726 727 728
}

void machine_restart(char *cmd)
{
	machine_ops.restart(cmd);
}

void machine_halt(void)
{
	machine_ops.halt();
}

729 730 731 732 733 734
#ifdef CONFIG_KEXEC
void machine_crash_shutdown(struct pt_regs *regs)
{
	machine_ops.crash_shutdown(regs);
}
#endif
735 736


737
#if defined(CONFIG_SMP)
738 739 740 741 742 743 744 745 746 747 748 749 750 751 752 753 754 755 756 757 758 759 760 761 762 763 764 765 766 767 768 769 770 771 772 773 774 775

/* This keeps a track of which one is crashing cpu. */
static int crashing_cpu;
static nmi_shootdown_cb shootdown_callback;

static atomic_t waiting_for_crash_ipi;

static int crash_nmi_callback(struct notifier_block *self,
			unsigned long val, void *data)
{
	int cpu;

	if (val != DIE_NMI_IPI)
		return NOTIFY_OK;

	cpu = raw_smp_processor_id();

	/* Don't do anything if this handler is invoked on crashing cpu.
	 * Otherwise, system will completely hang. Crashing cpu can get
	 * an NMI if system was initially booted with nmi_watchdog parameter.
	 */
	if (cpu == crashing_cpu)
		return NOTIFY_STOP;
	local_irq_disable();

	shootdown_callback(cpu, (struct die_args *)data);

	atomic_dec(&waiting_for_crash_ipi);
	/* Assume hlt works */
	halt();
	for (;;)
		cpu_relax();

	return 1;
}

static void smp_send_nmi_allbutself(void)
{
776
	apic->send_IPI_allbutself(NMI_VECTOR);
777 778 779 780
}

static struct notifier_block crash_nmi_nb = {
	.notifier_call = crash_nmi_callback,
781 782
	/* we want to be the first one called */
	.priority = NMI_LOCAL_HIGH_PRIOR+1,
783 784
};

785 786 787 788 789 790
/* Halt all other CPUs, calling the specified function on each of them
 *
 * This function can be used to halt all other CPUs on crash
 * or emergency reboot time. The function passed as parameter
 * will be called inside a NMI handler on all CPUs.
 */
791 792 793
void nmi_shootdown_cpus(nmi_shootdown_cb callback)
{
	unsigned long msecs;
794
	local_irq_disable();
795 796 797 798 799 800 801 802 803 804 805 806 807 808 809 810 811 812 813 814 815 816 817 818 819

	/* Make a note of crashing cpu. Will be used in NMI callback.*/
	crashing_cpu = safe_smp_processor_id();

	shootdown_callback = callback;

	atomic_set(&waiting_for_crash_ipi, num_online_cpus() - 1);
	/* Would it be better to replace the trap vector here? */
	if (register_die_notifier(&crash_nmi_nb))
		return;		/* return what? */
	/* Ensure the new callback function is set before sending
	 * out the NMI
	 */
	wmb();

	smp_send_nmi_allbutself();

	msecs = 1000; /* Wait at most a second for the other cpus to stop */
	while ((atomic_read(&waiting_for_crash_ipi) > 0) && msecs) {
		mdelay(1);
		msecs--;
	}

	/* Leave the nmi callback set */
}
820 821 822 823 824
#else /* !CONFIG_SMP */
void nmi_shootdown_cpus(nmi_shootdown_cb callback)
{
	/* No other CPUs to shoot down */
}
825
#endif