bugs.c 36.5 KB
Newer Older
1
// SPDX-License-Identifier: GPL-2.0
2 3 4 5 6 7 8 9 10 11 12
/*
 *  Copyright (C) 1994  Linus Torvalds
 *
 *  Cyrix stuff, June 1998 by:
 *	- Rafael R. Reilova (moved everything from head.S),
 *        <rreilova@ececs.uc.edu>
 *	- Channing Corn (tests & fixes),
 *	- Andrew D. Balsa (code cleanup).
 */
#include <linux/init.h>
#include <linux/utsname.h>
13
#include <linux/cpu.h>
14
#include <linux/module.h>
15 16
#include <linux/nospec.h>
#include <linux/prctl.h>
17
#include <linux/sched/smt.h>
18

19
#include <asm/spec-ctrl.h>
20
#include <asm/cmdline.h>
21
#include <asm/bugs.h>
22
#include <asm/processor.h>
D
Dave Jones 已提交
23
#include <asm/processor-flags.h>
24
#include <asm/fpu/internal.h>
25
#include <asm/msr.h>
26
#include <asm/vmx.h>
27 28
#include <asm/paravirt.h>
#include <asm/alternative.h>
29
#include <asm/pgtable.h>
L
Laura Abbott 已提交
30
#include <asm/set_memory.h>
31
#include <asm/intel-family.h>
32
#include <asm/e820/api.h>
33
#include <asm/hypervisor.h>
34

35
static void __init spectre_v2_select_mitigation(void);
36
static void __init ssb_select_mitigation(void);
37
static void __init l1tf_select_mitigation(void);
38
static void __init mds_select_mitigation(void);
39

40 41
/* The base value of the SPEC_CTRL MSR that always has to be preserved. */
u64 x86_spec_ctrl_base;
42
EXPORT_SYMBOL_GPL(x86_spec_ctrl_base);
43
static DEFINE_MUTEX(spec_ctrl_mutex);
44

45 46 47 48
/*
 * The vendor and possibly platform specific bits which can be modified in
 * x86_spec_ctrl_base.
 */
49
static u64 __ro_after_init x86_spec_ctrl_mask = SPEC_CTRL_IBRS;
50

51 52
/*
 * AMD specific MSR info for Speculative Store Bypass control.
53
 * x86_amd_ls_cfg_ssbd_mask is initialized in identify_boot_cpu().
54 55
 */
u64 __ro_after_init x86_amd_ls_cfg_base;
56
u64 __ro_after_init x86_amd_ls_cfg_ssbd_mask;
57

58 59
/* Control conditional STIPB in switch_to() */
DEFINE_STATIC_KEY_FALSE(switch_to_cond_stibp);
60 61 62 63
/* Control conditional IBPB in switch_mm() */
DEFINE_STATIC_KEY_FALSE(switch_mm_cond_ibpb);
/* Control unconditional IBPB in switch_mm() */
DEFINE_STATIC_KEY_FALSE(switch_mm_always_ibpb);
64

65 66
/* Control MDS CPU buffer clear before returning to user space */
DEFINE_STATIC_KEY_FALSE(mds_user_clear);
67
EXPORT_SYMBOL_GPL(mds_user_clear);
68 69 70
/* Control MDS CPU buffer clear before idling (halt, mwait) */
DEFINE_STATIC_KEY_FALSE(mds_idle_clear);
EXPORT_SYMBOL_GPL(mds_idle_clear);
71

72 73 74
void __init check_bugs(void)
{
	identify_boot_cpu();
75

76 77 78 79
	/*
	 * identify_boot_cpu() initialized SMT support information, let the
	 * core code know.
	 */
80
	cpu_smt_check_topology();
81

82 83 84 85 86
	if (!IS_ENABLED(CONFIG_SMP)) {
		pr_info("CPU: ");
		print_cpu_info(&boot_cpu_data);
	}

87 88
	/*
	 * Read the SPEC_CTRL MSR to account for reserved bits which may
89 90
	 * have unknown values. AMD64_LS_CFG MSR is cached in the early AMD
	 * init code as it is not enumerated and depends on the family.
91
	 */
92
	if (boot_cpu_has(X86_FEATURE_MSR_SPEC_CTRL))
93 94
		rdmsrl(MSR_IA32_SPEC_CTRL, x86_spec_ctrl_base);

95 96 97 98
	/* Allow STIBP in MSR_SPEC_CTRL if supported */
	if (boot_cpu_has(X86_FEATURE_STIBP))
		x86_spec_ctrl_mask |= SPEC_CTRL_STIBP;

99 100 101
	/* Select the proper spectre mitigation before patching alternatives */
	spectre_v2_select_mitigation();

102 103 104 105 106 107
	/*
	 * Select proper mitigation for any exposure to the Speculative Store
	 * Bypass vulnerability.
	 */
	ssb_select_mitigation();

108 109
	l1tf_select_mitigation();

110 111
	mds_select_mitigation();

112 113
	arch_smt_update();

114
#ifdef CONFIG_X86_32
115 116 117 118 119 120 121 122 123 124
	/*
	 * Check whether we are able to run this kernel safely on SMP.
	 *
	 * - i386 is no longer supported.
	 * - In order to run on anything without a TSC, we need to be
	 *   compiled for a i486.
	 */
	if (boot_cpu_data.x86 < 4)
		panic("Kernel requires i486+ for 'invlpg' and other features");

M
Miklos Vajna 已提交
125 126
	init_utsname()->machine[1] =
		'0' + (boot_cpu_data.x86 > 6 ? 6 : boot_cpu_data.x86);
127
	alternative_instructions();
128

129
	fpu__init_check_bugs();
130 131 132 133 134 135 136 137 138 139 140 141 142 143
#else /* CONFIG_X86_64 */
	alternative_instructions();

	/*
	 * Make sure the first 2MB area is not mapped by huge pages
	 * There are typically fixed size MTRRs in there and overlapping
	 * MTRRs into large pages causes slow downs.
	 *
	 * Right now we don't do that with gbpages because there seems
	 * very little benefit for that case.
	 */
	if (!direct_gbpages)
		set_memory_4k((unsigned long)__va(0), 1);
#endif
144
}
145

146 147
void
x86_virt_spec_ctrl(u64 guest_spec_ctrl, u64 guest_virt_spec_ctrl, bool setguest)
148
{
149
	u64 msrval, guestval, hostval = x86_spec_ctrl_base;
150
	struct thread_info *ti = current_thread_info();
151

152
	/* Is MSR_SPEC_CTRL implemented ? */
153
	if (static_cpu_has(X86_FEATURE_MSR_SPEC_CTRL)) {
154 155 156 157 158 159 160 161
		/*
		 * Restrict guest_spec_ctrl to supported values. Clear the
		 * modifiable bits in the host base value and or the
		 * modifiable bits from the guest value.
		 */
		guestval = hostval & ~x86_spec_ctrl_mask;
		guestval |= guest_spec_ctrl & x86_spec_ctrl_mask;

162
		/* SSBD controlled in MSR_SPEC_CTRL */
163 164
		if (static_cpu_has(X86_FEATURE_SPEC_CTRL_SSBD) ||
		    static_cpu_has(X86_FEATURE_AMD_SSBD))
165
			hostval |= ssbd_tif_to_spec_ctrl(ti->flags);
166

167 168 169 170
		/* Conditional STIBP enabled? */
		if (static_branch_unlikely(&switch_to_cond_stibp))
			hostval |= stibp_tif_to_spec_ctrl(ti->flags);

171 172 173
		if (hostval != guestval) {
			msrval = setguest ? guestval : hostval;
			wrmsrl(MSR_IA32_SPEC_CTRL, msrval);
174 175
		}
	}
176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203

	/*
	 * If SSBD is not handled in MSR_SPEC_CTRL on AMD, update
	 * MSR_AMD64_L2_CFG or MSR_VIRT_SPEC_CTRL if supported.
	 */
	if (!static_cpu_has(X86_FEATURE_LS_CFG_SSBD) &&
	    !static_cpu_has(X86_FEATURE_VIRT_SSBD))
		return;

	/*
	 * If the host has SSBD mitigation enabled, force it in the host's
	 * virtual MSR value. If its not permanently enabled, evaluate
	 * current's TIF_SSBD thread flag.
	 */
	if (static_cpu_has(X86_FEATURE_SPEC_STORE_BYPASS_DISABLE))
		hostval = SPEC_CTRL_SSBD;
	else
		hostval = ssbd_tif_to_spec_ctrl(ti->flags);

	/* Sanitize the guest value */
	guestval = guest_virt_spec_ctrl & SPEC_CTRL_SSBD;

	if (hostval != guestval) {
		unsigned long tif;

		tif = setguest ? ssbd_spec_ctrl_to_tif(guestval) :
				 ssbd_spec_ctrl_to_tif(hostval);

204
		speculation_ctrl_update(tif);
205
	}
206
}
207
EXPORT_SYMBOL_GPL(x86_virt_spec_ctrl);
208

209
static void x86_amd_ssb_disable(void)
210
{
211
	u64 msrval = x86_amd_ls_cfg_base | x86_amd_ls_cfg_ssbd_mask;
212

213 214 215
	if (boot_cpu_has(X86_FEATURE_VIRT_SSBD))
		wrmsrl(MSR_AMD64_VIRT_SPEC_CTRL, SPEC_CTRL_SSBD);
	else if (boot_cpu_has(X86_FEATURE_LS_CFG_SSBD))
216 217 218
		wrmsrl(MSR_AMD64_LS_CFG, msrval);
}

219 220 221
#undef pr_fmt
#define pr_fmt(fmt)	"MDS: " fmt

222
/* Default mitigation for MDS-affected CPUs */
223
static enum mds_mitigations mds_mitigation __ro_after_init = MDS_MITIGATION_FULL;
224
static bool mds_nosmt __ro_after_init = false;
225 226 227

static const char * const mds_strings[] = {
	[MDS_MITIGATION_OFF]	= "Vulnerable",
228 229
	[MDS_MITIGATION_FULL]	= "Mitigation: Clear CPU buffers",
	[MDS_MITIGATION_VMWERV]	= "Vulnerable: Clear CPU buffers attempted, no microcode",
230 231 232 233
};

static void __init mds_select_mitigation(void)
{
234
	if (!boot_cpu_has_bug(X86_BUG_MDS) || cpu_mitigations_off()) {
235 236 237 238 239
		mds_mitigation = MDS_MITIGATION_OFF;
		return;
	}

	if (mds_mitigation == MDS_MITIGATION_FULL) {
240 241
		if (!boot_cpu_has(X86_FEATURE_MD_CLEAR))
			mds_mitigation = MDS_MITIGATION_VMWERV;
242

243
		static_branch_enable(&mds_user_clear);
244

245 246
		if (!boot_cpu_has(X86_BUG_MSBDS_ONLY) &&
		    (mds_nosmt || cpu_mitigations_auto_nosmt()))
247
			cpu_smt_disable(false);
248
	}
249

250 251 252 253 254 255 256 257 258 259 260 261 262 263 264
	pr_info("%s\n", mds_strings[mds_mitigation]);
}

static int __init mds_cmdline(char *str)
{
	if (!boot_cpu_has_bug(X86_BUG_MDS))
		return 0;

	if (!str)
		return -EINVAL;

	if (!strcmp(str, "off"))
		mds_mitigation = MDS_MITIGATION_OFF;
	else if (!strcmp(str, "full"))
		mds_mitigation = MDS_MITIGATION_FULL;
265 266 267 268
	else if (!strcmp(str, "full,nosmt")) {
		mds_mitigation = MDS_MITIGATION_FULL;
		mds_nosmt = true;
	}
269 270 271 272 273

	return 0;
}
early_param("mds", mds_cmdline);

274 275 276 277 278 279
#undef pr_fmt
#define pr_fmt(fmt)     "Spectre V2 : " fmt

static enum spectre_v2_mitigation spectre_v2_enabled __ro_after_init =
	SPECTRE_V2_NONE;

280 281 282
static enum spectre_v2_user_mitigation spectre_v2_user __ro_after_init =
	SPECTRE_V2_USER_NONE;

283
#ifdef CONFIG_RETPOLINE
284 285
static bool spectre_v2_bad_module;

286 287 288 289 290
bool retpoline_module_ok(bool has_retpoline)
{
	if (spectre_v2_enabled == SPECTRE_V2_NONE || has_retpoline)
		return true;

291
	pr_err("System may be vulnerable to spectre v2\n");
292 293 294
	spectre_v2_bad_module = true;
	return false;
}
295 296 297 298 299 300 301

static inline const char *spectre_v2_module_string(void)
{
	return spectre_v2_bad_module ? " - vulnerable module loaded" : "";
}
#else
static inline const char *spectre_v2_module_string(void) { return ""; }
302
#endif
303 304 305 306 307 308 309 310

static inline bool match_option(const char *arg, int arglen, const char *opt)
{
	int len = strlen(opt);

	return len == arglen && !strncmp(arg, opt, len);
}

311 312 313 314 315 316 317 318 319 320
/* The kernel command line selection for spectre v2 */
enum spectre_v2_mitigation_cmd {
	SPECTRE_V2_CMD_NONE,
	SPECTRE_V2_CMD_AUTO,
	SPECTRE_V2_CMD_FORCE,
	SPECTRE_V2_CMD_RETPOLINE,
	SPECTRE_V2_CMD_RETPOLINE_GENERIC,
	SPECTRE_V2_CMD_RETPOLINE_AMD,
};

321 322 323 324
enum spectre_v2_user_cmd {
	SPECTRE_V2_USER_CMD_NONE,
	SPECTRE_V2_USER_CMD_AUTO,
	SPECTRE_V2_USER_CMD_FORCE,
325
	SPECTRE_V2_USER_CMD_PRCTL,
326
	SPECTRE_V2_USER_CMD_PRCTL_IBPB,
327
	SPECTRE_V2_USER_CMD_SECCOMP,
328
	SPECTRE_V2_USER_CMD_SECCOMP_IBPB,
329 330 331 332 333
};

static const char * const spectre_v2_user_strings[] = {
	[SPECTRE_V2_USER_NONE]		= "User space: Vulnerable",
	[SPECTRE_V2_USER_STRICT]	= "User space: Mitigation: STIBP protection",
334
	[SPECTRE_V2_USER_PRCTL]		= "User space: Mitigation: STIBP via prctl",
335
	[SPECTRE_V2_USER_SECCOMP]	= "User space: Mitigation: STIBP via seccomp and prctl",
336 337 338 339 340 341
};

static const struct {
	const char			*option;
	enum spectre_v2_user_cmd	cmd;
	bool				secure;
342
} v2_user_options[] __initconst = {
343 344 345 346 347 348 349
	{ "auto",		SPECTRE_V2_USER_CMD_AUTO,		false },
	{ "off",		SPECTRE_V2_USER_CMD_NONE,		false },
	{ "on",			SPECTRE_V2_USER_CMD_FORCE,		true  },
	{ "prctl",		SPECTRE_V2_USER_CMD_PRCTL,		false },
	{ "prctl,ibpb",		SPECTRE_V2_USER_CMD_PRCTL_IBPB,		false },
	{ "seccomp",		SPECTRE_V2_USER_CMD_SECCOMP,		false },
	{ "seccomp,ibpb",	SPECTRE_V2_USER_CMD_SECCOMP_IBPB,	false },
350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394
};

static void __init spec_v2_user_print_cond(const char *reason, bool secure)
{
	if (boot_cpu_has_bug(X86_BUG_SPECTRE_V2) != secure)
		pr_info("spectre_v2_user=%s forced on command line.\n", reason);
}

static enum spectre_v2_user_cmd __init
spectre_v2_parse_user_cmdline(enum spectre_v2_mitigation_cmd v2_cmd)
{
	char arg[20];
	int ret, i;

	switch (v2_cmd) {
	case SPECTRE_V2_CMD_NONE:
		return SPECTRE_V2_USER_CMD_NONE;
	case SPECTRE_V2_CMD_FORCE:
		return SPECTRE_V2_USER_CMD_FORCE;
	default:
		break;
	}

	ret = cmdline_find_option(boot_command_line, "spectre_v2_user",
				  arg, sizeof(arg));
	if (ret < 0)
		return SPECTRE_V2_USER_CMD_AUTO;

	for (i = 0; i < ARRAY_SIZE(v2_user_options); i++) {
		if (match_option(arg, ret, v2_user_options[i].option)) {
			spec_v2_user_print_cond(v2_user_options[i].option,
						v2_user_options[i].secure);
			return v2_user_options[i].cmd;
		}
	}

	pr_err("Unknown user space protection option (%s). Switching to AUTO select\n", arg);
	return SPECTRE_V2_USER_CMD_AUTO;
}

static void __init
spectre_v2_user_select_mitigation(enum spectre_v2_mitigation_cmd v2_cmd)
{
	enum spectre_v2_user_mitigation mode = SPECTRE_V2_USER_NONE;
	bool smt_possible = IS_ENABLED(CONFIG_SMP);
395
	enum spectre_v2_user_cmd cmd;
396 397 398 399 400 401 402 403

	if (!boot_cpu_has(X86_FEATURE_IBPB) && !boot_cpu_has(X86_FEATURE_STIBP))
		return;

	if (cpu_smt_control == CPU_SMT_FORCE_DISABLED ||
	    cpu_smt_control == CPU_SMT_NOT_SUPPORTED)
		smt_possible = false;

404 405
	cmd = spectre_v2_parse_user_cmdline(v2_cmd);
	switch (cmd) {
406 407 408 409 410
	case SPECTRE_V2_USER_CMD_NONE:
		goto set_mode;
	case SPECTRE_V2_USER_CMD_FORCE:
		mode = SPECTRE_V2_USER_STRICT;
		break;
411
	case SPECTRE_V2_USER_CMD_PRCTL:
412
	case SPECTRE_V2_USER_CMD_PRCTL_IBPB:
413 414
		mode = SPECTRE_V2_USER_PRCTL;
		break;
415 416
	case SPECTRE_V2_USER_CMD_AUTO:
	case SPECTRE_V2_USER_CMD_SECCOMP:
417
	case SPECTRE_V2_USER_CMD_SECCOMP_IBPB:
418 419 420 421 422
		if (IS_ENABLED(CONFIG_SECCOMP))
			mode = SPECTRE_V2_USER_SECCOMP;
		else
			mode = SPECTRE_V2_USER_PRCTL;
		break;
423 424 425 426 427
	}

	/* Initialize Indirect Branch Prediction Barrier */
	if (boot_cpu_has(X86_FEATURE_IBPB)) {
		setup_force_cpu_cap(X86_FEATURE_USE_IBPB);
428

429 430 431 432
		switch (cmd) {
		case SPECTRE_V2_USER_CMD_FORCE:
		case SPECTRE_V2_USER_CMD_PRCTL_IBPB:
		case SPECTRE_V2_USER_CMD_SECCOMP_IBPB:
433 434
			static_branch_enable(&switch_mm_always_ibpb);
			break;
435 436 437
		case SPECTRE_V2_USER_CMD_PRCTL:
		case SPECTRE_V2_USER_CMD_AUTO:
		case SPECTRE_V2_USER_CMD_SECCOMP:
438 439
			static_branch_enable(&switch_mm_cond_ibpb);
			break;
440 441 442 443 444
		default:
			break;
		}

		pr_info("mitigation: Enabling %s Indirect Branch Prediction Barrier\n",
445 446
			static_key_enabled(&switch_mm_always_ibpb) ?
			"always-on" : "conditional");
447 448 449 450 451 452
	}

	/* If enhanced IBRS is enabled no STIPB required */
	if (spectre_v2_enabled == SPECTRE_V2_IBRS_ENHANCED)
		return;

453 454 455 456 457 458
	/*
	 * If SMT is not possible or STIBP is not available clear the STIPB
	 * mode.
	 */
	if (!smt_possible || !boot_cpu_has(X86_FEATURE_STIBP))
		mode = SPECTRE_V2_USER_NONE;
459 460 461 462 463 464 465
set_mode:
	spectre_v2_user = mode;
	/* Only print the STIBP mode when SMT possible */
	if (smt_possible)
		pr_info("%s\n", spectre_v2_user_strings[mode]);
}

466
static const char * const spectre_v2_strings[] = {
467 468 469 470 471 472
	[SPECTRE_V2_NONE]			= "Vulnerable",
	[SPECTRE_V2_RETPOLINE_GENERIC]		= "Mitigation: Full generic retpoline",
	[SPECTRE_V2_RETPOLINE_AMD]		= "Mitigation: Full AMD retpoline",
	[SPECTRE_V2_IBRS_ENHANCED]		= "Mitigation: Enhanced IBRS",
};

473 474 475 476
static const struct {
	const char *option;
	enum spectre_v2_mitigation_cmd cmd;
	bool secure;
477
} mitigation_options[] __initconst = {
478 479 480 481 482 483
	{ "off",		SPECTRE_V2_CMD_NONE,		  false },
	{ "on",			SPECTRE_V2_CMD_FORCE,		  true  },
	{ "retpoline",		SPECTRE_V2_CMD_RETPOLINE,	  false },
	{ "retpoline,amd",	SPECTRE_V2_CMD_RETPOLINE_AMD,	  false },
	{ "retpoline,generic",	SPECTRE_V2_CMD_RETPOLINE_GENERIC, false },
	{ "auto",		SPECTRE_V2_CMD_AUTO,		  false },
484 485
};

486
static void __init spec_v2_print_cond(const char *reason, bool secure)
487
{
488
	if (boot_cpu_has_bug(X86_BUG_SPECTRE_V2) != secure)
489 490 491
		pr_info("%s selected on command line.\n", reason);
}

492 493
static enum spectre_v2_mitigation_cmd __init spectre_v2_parse_cmdline(void)
{
494
	enum spectre_v2_mitigation_cmd cmd = SPECTRE_V2_CMD_AUTO;
495
	char arg[20];
496 497
	int ret, i;

498 499
	if (cmdline_find_option_bool(boot_command_line, "nospectre_v2") ||
	    cpu_mitigations_off())
500 501
		return SPECTRE_V2_CMD_NONE;

502 503 504 505 506 507 508 509 510 511 512 513 514 515
	ret = cmdline_find_option(boot_command_line, "spectre_v2", arg, sizeof(arg));
	if (ret < 0)
		return SPECTRE_V2_CMD_AUTO;

	for (i = 0; i < ARRAY_SIZE(mitigation_options); i++) {
		if (!match_option(arg, ret, mitigation_options[i].option))
			continue;
		cmd = mitigation_options[i].cmd;
		break;
	}

	if (i >= ARRAY_SIZE(mitigation_options)) {
		pr_err("unknown option (%s). Switching to AUTO select\n", arg);
		return SPECTRE_V2_CMD_AUTO;
516 517
	}

518 519 520 521
	if ((cmd == SPECTRE_V2_CMD_RETPOLINE ||
	     cmd == SPECTRE_V2_CMD_RETPOLINE_AMD ||
	     cmd == SPECTRE_V2_CMD_RETPOLINE_GENERIC) &&
	    !IS_ENABLED(CONFIG_RETPOLINE)) {
522
		pr_err("%s selected but not compiled in. Switching to AUTO select\n", mitigation_options[i].option);
523
		return SPECTRE_V2_CMD_AUTO;
524 525 526 527 528 529 530 531
	}

	if (cmd == SPECTRE_V2_CMD_RETPOLINE_AMD &&
	    boot_cpu_data.x86_vendor != X86_VENDOR_AMD) {
		pr_err("retpoline,amd selected but CPU is not AMD. Switching to AUTO select\n");
		return SPECTRE_V2_CMD_AUTO;
	}

532 533
	spec_v2_print_cond(mitigation_options[i].option,
			   mitigation_options[i].secure);
534
	return cmd;
535 536 537 538 539 540 541 542 543 544 545 546 547 548 549 550 551 552 553 554 555
}

static void __init spectre_v2_select_mitigation(void)
{
	enum spectre_v2_mitigation_cmd cmd = spectre_v2_parse_cmdline();
	enum spectre_v2_mitigation mode = SPECTRE_V2_NONE;

	/*
	 * If the CPU is not affected and the command line mode is NONE or AUTO
	 * then nothing to do.
	 */
	if (!boot_cpu_has_bug(X86_BUG_SPECTRE_V2) &&
	    (cmd == SPECTRE_V2_CMD_NONE || cmd == SPECTRE_V2_CMD_AUTO))
		return;

	switch (cmd) {
	case SPECTRE_V2_CMD_NONE:
		return;

	case SPECTRE_V2_CMD_FORCE:
	case SPECTRE_V2_CMD_AUTO:
556 557 558 559 560 561 562
		if (boot_cpu_has(X86_FEATURE_IBRS_ENHANCED)) {
			mode = SPECTRE_V2_IBRS_ENHANCED;
			/* Force it so VMEXIT will restore correctly */
			x86_spec_ctrl_base |= SPEC_CTRL_IBRS;
			wrmsrl(MSR_IA32_SPEC_CTRL, x86_spec_ctrl_base);
			goto specv2_set_mode;
		}
563 564 565
		if (IS_ENABLED(CONFIG_RETPOLINE))
			goto retpoline_auto;
		break;
566 567 568 569 570 571 572 573 574 575 576 577 578
	case SPECTRE_V2_CMD_RETPOLINE_AMD:
		if (IS_ENABLED(CONFIG_RETPOLINE))
			goto retpoline_amd;
		break;
	case SPECTRE_V2_CMD_RETPOLINE_GENERIC:
		if (IS_ENABLED(CONFIG_RETPOLINE))
			goto retpoline_generic;
		break;
	case SPECTRE_V2_CMD_RETPOLINE:
		if (IS_ENABLED(CONFIG_RETPOLINE))
			goto retpoline_auto;
		break;
	}
579
	pr_err("Spectre mitigation: kernel not compiled with retpoline; no mitigation available!");
580 581 582 583 584 585
	return;

retpoline_auto:
	if (boot_cpu_data.x86_vendor == X86_VENDOR_AMD) {
	retpoline_amd:
		if (!boot_cpu_has(X86_FEATURE_LFENCE_RDTSC)) {
586
			pr_err("Spectre mitigation: LFENCE not serializing, switching to generic retpoline\n");
587 588
			goto retpoline_generic;
		}
589
		mode = SPECTRE_V2_RETPOLINE_AMD;
590 591 592 593
		setup_force_cpu_cap(X86_FEATURE_RETPOLINE_AMD);
		setup_force_cpu_cap(X86_FEATURE_RETPOLINE);
	} else {
	retpoline_generic:
594
		mode = SPECTRE_V2_RETPOLINE_GENERIC;
595 596 597
		setup_force_cpu_cap(X86_FEATURE_RETPOLINE);
	}

598
specv2_set_mode:
599 600
	spectre_v2_enabled = mode;
	pr_info("%s\n", spectre_v2_strings[mode]);
601 602

	/*
603 604 605
	 * If spectre v2 protection has been enabled, unconditionally fill
	 * RSB during a context switch; this protects against two independent
	 * issues:
606
	 *
607 608
	 *	- RSB underflow (and switch to BTB) on Skylake+
	 *	- SpectreRSB variant of spectre v2 on X86_BUG_SPECTRE_V2 CPUs
609
	 */
610 611
	setup_force_cpu_cap(X86_FEATURE_RSB_CTXSW);
	pr_info("Spectre v2 / SpectreRSB mitigation: Filling RSB on context switch\n");
612

613 614
	/*
	 * Retpoline means the kernel is safe because it has no indirect
615 616 617 618 619 620 621 622
	 * branches. Enhanced IBRS protects firmware too, so, enable restricted
	 * speculation around firmware calls only when Enhanced IBRS isn't
	 * supported.
	 *
	 * Use "mode" to check Enhanced IBRS instead of boot_cpu_has(), because
	 * the user might select retpoline on the kernel command line and if
	 * the CPU supports Enhanced IBRS, kernel might un-intentionally not
	 * enable IBRS around firmware calls.
623
	 */
624
	if (boot_cpu_has(X86_FEATURE_IBRS) && mode != SPECTRE_V2_IBRS_ENHANCED) {
625 626 627
		setup_force_cpu_cap(X86_FEATURE_USE_IBRS_FW);
		pr_info("Enabling Restricted Speculation for firmware calls\n");
	}
628

629 630
	/* Set up IBPB and STIBP depending on the general spectre V2 command */
	spectre_v2_user_select_mitigation(cmd);
631 632
}

633
static void update_stibp_msr(void * __unused)
634
{
635
	wrmsrl(MSR_IA32_SPEC_CTRL, x86_spec_ctrl_base);
636 637
}

638 639
/* Update x86_spec_ctrl_base in case SMT state changed. */
static void update_stibp_strict(void)
640
{
641 642 643 644 645 646 647 648 649 650 651 652
	u64 mask = x86_spec_ctrl_base & ~SPEC_CTRL_STIBP;

	if (sched_smt_active())
		mask |= SPEC_CTRL_STIBP;

	if (mask == x86_spec_ctrl_base)
		return;

	pr_info("Update user space SMT mitigation: STIBP %s\n",
		mask & SPEC_CTRL_STIBP ? "always-on" : "off");
	x86_spec_ctrl_base = mask;
	on_each_cpu(update_stibp_msr, NULL, 1);
653 654
}

655 656 657 658 659 660 661 662 663
/* Update the static key controlling the evaluation of TIF_SPEC_IB */
static void update_indir_branch_cond(void)
{
	if (sched_smt_active())
		static_branch_enable(&switch_to_cond_stibp);
	else
		static_branch_disable(&switch_to_cond_stibp);
}

664 665 666
#undef pr_fmt
#define pr_fmt(fmt) fmt

667 668 669 670 671 672 673 674 675 676 677 678 679 680 681 682 683 684 685 686
/* Update the static key controlling the MDS CPU buffer clear in idle */
static void update_mds_branch_idle(void)
{
	/*
	 * Enable the idle clearing if SMT is active on CPUs which are
	 * affected only by MSBDS and not any other MDS variant.
	 *
	 * The other variants cannot be mitigated when SMT is enabled, so
	 * clearing the buffers on idle just to prevent the Store Buffer
	 * repartitioning leak would be a window dressing exercise.
	 */
	if (!boot_cpu_has_bug(X86_BUG_MSBDS_ONLY))
		return;

	if (sched_smt_active())
		static_branch_enable(&mds_idle_clear);
	else
		static_branch_disable(&mds_idle_clear);
}

687 688
#define MDS_MSG_SMT "MDS CPU bug present and SMT on, data leak possible. See https://www.kernel.org/doc/html/latest/admin-guide/hw-vuln/mds.html for more details.\n"

689 690
void arch_smt_update(void)
{
691 692
	/* Enhanced IBRS implies STIBP. No update required. */
	if (spectre_v2_enabled == SPECTRE_V2_IBRS_ENHANCED)
693 694 695 696
		return;

	mutex_lock(&spec_ctrl_mutex);

697 698 699 700 701 702
	switch (spectre_v2_user) {
	case SPECTRE_V2_USER_NONE:
		break;
	case SPECTRE_V2_USER_STRICT:
		update_stibp_strict();
		break;
703
	case SPECTRE_V2_USER_PRCTL:
704
	case SPECTRE_V2_USER_SECCOMP:
705
		update_indir_branch_cond();
706
		break;
707
	}
708

709 710 711
	switch (mds_mitigation) {
	case MDS_MITIGATION_FULL:
	case MDS_MITIGATION_VMWERV:
712 713
		if (sched_smt_active() && !boot_cpu_has(X86_BUG_MSBDS_ONLY))
			pr_warn_once(MDS_MSG_SMT);
714
		update_mds_branch_idle();
715 716 717 718
		break;
	case MDS_MITIGATION_OFF:
		break;
	}
719

720 721 722
	mutex_unlock(&spec_ctrl_mutex);
}

723 724 725
#undef pr_fmt
#define pr_fmt(fmt)	"Speculative Store Bypass: " fmt

726
static enum ssb_mitigation ssb_mode __ro_after_init = SPEC_STORE_BYPASS_NONE;
727 728 729 730 731 732

/* The kernel command line selection */
enum ssb_mitigation_cmd {
	SPEC_STORE_BYPASS_CMD_NONE,
	SPEC_STORE_BYPASS_CMD_AUTO,
	SPEC_STORE_BYPASS_CMD_ON,
733
	SPEC_STORE_BYPASS_CMD_PRCTL,
734
	SPEC_STORE_BYPASS_CMD_SECCOMP,
735 736
};

737
static const char * const ssb_strings[] = {
738
	[SPEC_STORE_BYPASS_NONE]	= "Vulnerable",
739
	[SPEC_STORE_BYPASS_DISABLE]	= "Mitigation: Speculative Store Bypass disabled",
740 741
	[SPEC_STORE_BYPASS_PRCTL]	= "Mitigation: Speculative Store Bypass disabled via prctl",
	[SPEC_STORE_BYPASS_SECCOMP]	= "Mitigation: Speculative Store Bypass disabled via prctl and seccomp",
742 743 744 745 746
};

static const struct {
	const char *option;
	enum ssb_mitigation_cmd cmd;
747
} ssb_mitigation_options[]  __initconst = {
748 749 750 751 752
	{ "auto",	SPEC_STORE_BYPASS_CMD_AUTO },    /* Platform decides */
	{ "on",		SPEC_STORE_BYPASS_CMD_ON },      /* Disable Speculative Store Bypass */
	{ "off",	SPEC_STORE_BYPASS_CMD_NONE },    /* Don't touch Speculative Store Bypass */
	{ "prctl",	SPEC_STORE_BYPASS_CMD_PRCTL },   /* Disable Speculative Store Bypass via prctl */
	{ "seccomp",	SPEC_STORE_BYPASS_CMD_SECCOMP }, /* Disable Speculative Store Bypass via prctl and seccomp */
753 754 755 756 757 758 759 760
};

static enum ssb_mitigation_cmd __init ssb_parse_cmdline(void)
{
	enum ssb_mitigation_cmd cmd = SPEC_STORE_BYPASS_CMD_AUTO;
	char arg[20];
	int ret, i;

761 762
	if (cmdline_find_option_bool(boot_command_line, "nospec_store_bypass_disable") ||
	    cpu_mitigations_off()) {
763 764 765 766 767 768 769 770 771 772 773 774 775 776 777 778 779 780 781 782 783 784 785 786
		return SPEC_STORE_BYPASS_CMD_NONE;
	} else {
		ret = cmdline_find_option(boot_command_line, "spec_store_bypass_disable",
					  arg, sizeof(arg));
		if (ret < 0)
			return SPEC_STORE_BYPASS_CMD_AUTO;

		for (i = 0; i < ARRAY_SIZE(ssb_mitigation_options); i++) {
			if (!match_option(arg, ret, ssb_mitigation_options[i].option))
				continue;

			cmd = ssb_mitigation_options[i].cmd;
			break;
		}

		if (i >= ARRAY_SIZE(ssb_mitigation_options)) {
			pr_err("unknown option (%s). Switching to AUTO select\n", arg);
			return SPEC_STORE_BYPASS_CMD_AUTO;
		}
	}

	return cmd;
}

787
static enum ssb_mitigation __init __ssb_select_mitigation(void)
788 789 790 791
{
	enum ssb_mitigation mode = SPEC_STORE_BYPASS_NONE;
	enum ssb_mitigation_cmd cmd;

792
	if (!boot_cpu_has(X86_FEATURE_SSBD))
793 794 795 796 797 798 799 800 801 802
		return mode;

	cmd = ssb_parse_cmdline();
	if (!boot_cpu_has_bug(X86_BUG_SPEC_STORE_BYPASS) &&
	    (cmd == SPEC_STORE_BYPASS_CMD_NONE ||
	     cmd == SPEC_STORE_BYPASS_CMD_AUTO))
		return mode;

	switch (cmd) {
	case SPEC_STORE_BYPASS_CMD_AUTO:
803 804 805 806 807 808 809 810 811
	case SPEC_STORE_BYPASS_CMD_SECCOMP:
		/*
		 * Choose prctl+seccomp as the default mode if seccomp is
		 * enabled.
		 */
		if (IS_ENABLED(CONFIG_SECCOMP))
			mode = SPEC_STORE_BYPASS_SECCOMP;
		else
			mode = SPEC_STORE_BYPASS_PRCTL;
812
		break;
813 814 815
	case SPEC_STORE_BYPASS_CMD_ON:
		mode = SPEC_STORE_BYPASS_DISABLE;
		break;
816 817 818
	case SPEC_STORE_BYPASS_CMD_PRCTL:
		mode = SPEC_STORE_BYPASS_PRCTL;
		break;
819 820 821 822
	case SPEC_STORE_BYPASS_CMD_NONE:
		break;
	}

823 824 825 826 827 828 829 830 831 832
	/*
	 * If SSBD is controlled by the SPEC_CTRL MSR, then set the proper
	 * bit in the mask to allow guests to use the mitigation even in the
	 * case where the host does not enable it.
	 */
	if (static_cpu_has(X86_FEATURE_SPEC_CTRL_SSBD) ||
	    static_cpu_has(X86_FEATURE_AMD_SSBD)) {
		x86_spec_ctrl_mask |= SPEC_CTRL_SSBD;
	}

833 834 835
	/*
	 * We have three CPU feature flags that are in play here:
	 *  - X86_BUG_SPEC_STORE_BYPASS - CPU is susceptible.
836
	 *  - X86_FEATURE_SSBD - CPU is able to turn off speculative store bypass
837 838
	 *  - X86_FEATURE_SPEC_STORE_BYPASS_DISABLE - engage the mitigation
	 */
839
	if (mode == SPEC_STORE_BYPASS_DISABLE) {
840
		setup_force_cpu_cap(X86_FEATURE_SPEC_STORE_BYPASS_DISABLE);
841
		/*
842 843
		 * Intel uses the SPEC CTRL MSR Bit(2) for this, while AMD may
		 * use a completely different MSR and bit dependent on family.
844
		 */
845 846
		if (!static_cpu_has(X86_FEATURE_SPEC_CTRL_SSBD) &&
		    !static_cpu_has(X86_FEATURE_AMD_SSBD)) {
847
			x86_amd_ssb_disable();
848
		} else {
849
			x86_spec_ctrl_base |= SPEC_CTRL_SSBD;
850
			wrmsrl(MSR_IA32_SPEC_CTRL, x86_spec_ctrl_base);
851 852 853
		}
	}

854 855 856
	return mode;
}

857
static void ssb_select_mitigation(void)
858 859 860 861 862 863 864
{
	ssb_mode = __ssb_select_mitigation();

	if (boot_cpu_has_bug(X86_BUG_SPEC_STORE_BYPASS))
		pr_info("%s\n", ssb_strings[ssb_mode]);
}

865
#undef pr_fmt
866
#define pr_fmt(fmt)     "Speculation prctl: " fmt
867

868
static void task_update_spec_tif(struct task_struct *tsk)
869
{
870 871
	/* Force the update of the real TIF bits */
	set_tsk_thread_flag(tsk, TIF_SPEC_FORCE_UPDATE);
872 873 874 875 876 877 878 879 880

	/*
	 * Immediately update the speculation control MSRs for the current
	 * task, but for a non-current task delay setting the CPU
	 * mitigation until it is scheduled next.
	 *
	 * This can only happen for SECCOMP mitigation. For PRCTL it's
	 * always the current task.
	 */
881
	if (tsk == current)
882 883 884 885 886
		speculation_ctrl_update_current();
}

static int ssb_prctl_set(struct task_struct *task, unsigned long ctrl)
{
887 888
	if (ssb_mode != SPEC_STORE_BYPASS_PRCTL &&
	    ssb_mode != SPEC_STORE_BYPASS_SECCOMP)
889 890
		return -ENXIO;

891 892 893 894 895 896
	switch (ctrl) {
	case PR_SPEC_ENABLE:
		/* If speculation is force disabled, enable is not allowed */
		if (task_spec_ssb_force_disable(task))
			return -EPERM;
		task_clear_spec_ssb_disable(task);
897
		task_update_spec_tif(task);
898 899 900
		break;
	case PR_SPEC_DISABLE:
		task_set_spec_ssb_disable(task);
901
		task_update_spec_tif(task);
902 903 904 905
		break;
	case PR_SPEC_FORCE_DISABLE:
		task_set_spec_ssb_disable(task);
		task_set_spec_ssb_force_disable(task);
906
		task_update_spec_tif(task);
907 908 909 910
		break;
	default:
		return -ERANGE;
	}
911 912 913
	return 0;
}

914 915 916 917 918 919 920 921 922 923 924 925 926 927 928 929 930 931 932 933 934 935 936 937 938 939 940 941 942 943 944 945 946 947 948 949
static int ib_prctl_set(struct task_struct *task, unsigned long ctrl)
{
	switch (ctrl) {
	case PR_SPEC_ENABLE:
		if (spectre_v2_user == SPECTRE_V2_USER_NONE)
			return 0;
		/*
		 * Indirect branch speculation is always disabled in strict
		 * mode.
		 */
		if (spectre_v2_user == SPECTRE_V2_USER_STRICT)
			return -EPERM;
		task_clear_spec_ib_disable(task);
		task_update_spec_tif(task);
		break;
	case PR_SPEC_DISABLE:
	case PR_SPEC_FORCE_DISABLE:
		/*
		 * Indirect branch speculation is always allowed when
		 * mitigation is force disabled.
		 */
		if (spectre_v2_user == SPECTRE_V2_USER_NONE)
			return -EPERM;
		if (spectre_v2_user == SPECTRE_V2_USER_STRICT)
			return 0;
		task_set_spec_ib_disable(task);
		if (ctrl == PR_SPEC_FORCE_DISABLE)
			task_set_spec_ib_force_disable(task);
		task_update_spec_tif(task);
		break;
	default:
		return -ERANGE;
	}
	return 0;
}

950 951 952 953 954 955
int arch_prctl_spec_ctrl_set(struct task_struct *task, unsigned long which,
			     unsigned long ctrl)
{
	switch (which) {
	case PR_SPEC_STORE_BYPASS:
		return ssb_prctl_set(task, ctrl);
956 957
	case PR_SPEC_INDIRECT_BRANCH:
		return ib_prctl_set(task, ctrl);
958 959 960 961 962 963 964 965
	default:
		return -ENODEV;
	}
}

#ifdef CONFIG_SECCOMP
void arch_seccomp_spec_mitigate(struct task_struct *task)
{
966 967
	if (ssb_mode == SPEC_STORE_BYPASS_SECCOMP)
		ssb_prctl_set(task, PR_SPEC_FORCE_DISABLE);
968 969
	if (spectre_v2_user == SPECTRE_V2_USER_SECCOMP)
		ib_prctl_set(task, PR_SPEC_FORCE_DISABLE);
970 971 972
}
#endif

973
static int ssb_prctl_get(struct task_struct *task)
974 975 976 977
{
	switch (ssb_mode) {
	case SPEC_STORE_BYPASS_DISABLE:
		return PR_SPEC_DISABLE;
978
	case SPEC_STORE_BYPASS_SECCOMP:
979
	case SPEC_STORE_BYPASS_PRCTL:
980 981 982
		if (task_spec_ssb_force_disable(task))
			return PR_SPEC_PRCTL | PR_SPEC_FORCE_DISABLE;
		if (task_spec_ssb_disable(task))
983 984 985 986 987 988 989 990 991
			return PR_SPEC_PRCTL | PR_SPEC_DISABLE;
		return PR_SPEC_PRCTL | PR_SPEC_ENABLE;
	default:
		if (boot_cpu_has_bug(X86_BUG_SPEC_STORE_BYPASS))
			return PR_SPEC_ENABLE;
		return PR_SPEC_NOT_AFFECTED;
	}
}

992 993 994 995 996 997 998 999 1000
static int ib_prctl_get(struct task_struct *task)
{
	if (!boot_cpu_has_bug(X86_BUG_SPECTRE_V2))
		return PR_SPEC_NOT_AFFECTED;

	switch (spectre_v2_user) {
	case SPECTRE_V2_USER_NONE:
		return PR_SPEC_ENABLE;
	case SPECTRE_V2_USER_PRCTL:
1001
	case SPECTRE_V2_USER_SECCOMP:
1002 1003 1004 1005 1006 1007 1008 1009 1010 1011 1012 1013
		if (task_spec_ib_force_disable(task))
			return PR_SPEC_PRCTL | PR_SPEC_FORCE_DISABLE;
		if (task_spec_ib_disable(task))
			return PR_SPEC_PRCTL | PR_SPEC_DISABLE;
		return PR_SPEC_PRCTL | PR_SPEC_ENABLE;
	case SPECTRE_V2_USER_STRICT:
		return PR_SPEC_DISABLE;
	default:
		return PR_SPEC_NOT_AFFECTED;
	}
}

1014
int arch_prctl_spec_ctrl_get(struct task_struct *task, unsigned long which)
1015 1016 1017
{
	switch (which) {
	case PR_SPEC_STORE_BYPASS:
1018
		return ssb_prctl_get(task);
1019 1020
	case PR_SPEC_INDIRECT_BRANCH:
		return ib_prctl_get(task);
1021 1022 1023 1024 1025
	default:
		return -ENODEV;
	}
}

1026 1027
void x86_spec_ctrl_setup_ap(void)
{
1028
	if (boot_cpu_has(X86_FEATURE_MSR_SPEC_CTRL))
1029
		wrmsrl(MSR_IA32_SPEC_CTRL, x86_spec_ctrl_base);
1030 1031

	if (ssb_mode == SPEC_STORE_BYPASS_DISABLE)
1032
		x86_amd_ssb_disable();
1033 1034
}

1035 1036
#undef pr_fmt
#define pr_fmt(fmt)	"L1TF: " fmt
1037

1038 1039
/* Default mitigation for L1TF-affected CPUs */
enum l1tf_mitigations l1tf_mitigation __ro_after_init = L1TF_MITIGATION_FLUSH;
1040
#if IS_ENABLED(CONFIG_KVM_INTEL)
1041
EXPORT_SYMBOL_GPL(l1tf_mitigation);
1042
#endif
1043
enum vmx_l1d_flush_state l1tf_vmx_mitigation = VMENTER_L1D_FLUSH_AUTO;
1044 1045
EXPORT_SYMBOL_GPL(l1tf_vmx_mitigation);

1046 1047 1048 1049 1050 1051 1052 1053 1054 1055 1056 1057 1058 1059 1060 1061 1062 1063 1064 1065 1066 1067 1068 1069 1070 1071 1072 1073 1074 1075 1076 1077 1078 1079 1080 1081 1082 1083 1084
/*
 * These CPUs all support 44bits physical address space internally in the
 * cache but CPUID can report a smaller number of physical address bits.
 *
 * The L1TF mitigation uses the top most address bit for the inversion of
 * non present PTEs. When the installed memory reaches into the top most
 * address bit due to memory holes, which has been observed on machines
 * which report 36bits physical address bits and have 32G RAM installed,
 * then the mitigation range check in l1tf_select_mitigation() triggers.
 * This is a false positive because the mitigation is still possible due to
 * the fact that the cache uses 44bit internally. Use the cache bits
 * instead of the reported physical bits and adjust them on the affected
 * machines to 44bit if the reported bits are less than 44.
 */
static void override_cache_bits(struct cpuinfo_x86 *c)
{
	if (c->x86 != 6)
		return;

	switch (c->x86_model) {
	case INTEL_FAM6_NEHALEM:
	case INTEL_FAM6_WESTMERE:
	case INTEL_FAM6_SANDYBRIDGE:
	case INTEL_FAM6_IVYBRIDGE:
	case INTEL_FAM6_HASWELL_CORE:
	case INTEL_FAM6_HASWELL_ULT:
	case INTEL_FAM6_HASWELL_GT3E:
	case INTEL_FAM6_BROADWELL_CORE:
	case INTEL_FAM6_BROADWELL_GT3E:
	case INTEL_FAM6_SKYLAKE_MOBILE:
	case INTEL_FAM6_SKYLAKE_DESKTOP:
	case INTEL_FAM6_KABYLAKE_MOBILE:
	case INTEL_FAM6_KABYLAKE_DESKTOP:
		if (c->x86_cache_bits < 44)
			c->x86_cache_bits = 44;
		break;
	}
}

1085 1086 1087 1088 1089 1090 1091
static void __init l1tf_select_mitigation(void)
{
	u64 half_pa;

	if (!boot_cpu_has_bug(X86_BUG_L1TF))
		return;

1092 1093 1094 1095 1096
	if (cpu_mitigations_off())
		l1tf_mitigation = L1TF_MITIGATION_OFF;
	else if (cpu_mitigations_auto_nosmt())
		l1tf_mitigation = L1TF_MITIGATION_FLUSH_NOSMT;

1097 1098
	override_cache_bits(&boot_cpu_data);

1099 1100 1101 1102 1103 1104 1105 1106 1107 1108 1109 1110 1111 1112
	switch (l1tf_mitigation) {
	case L1TF_MITIGATION_OFF:
	case L1TF_MITIGATION_FLUSH_NOWARN:
	case L1TF_MITIGATION_FLUSH:
		break;
	case L1TF_MITIGATION_FLUSH_NOSMT:
	case L1TF_MITIGATION_FULL:
		cpu_smt_disable(false);
		break;
	case L1TF_MITIGATION_FULL_FORCE:
		cpu_smt_disable(true);
		break;
	}

1113 1114 1115 1116 1117 1118
#if CONFIG_PGTABLE_LEVELS == 2
	pr_warn("Kernel not compiled for PAE. No mitigation for L1TF\n");
	return;
#endif

	half_pa = (u64)l1tf_pfn_limit() << PAGE_SHIFT;
1119 1120
	if (l1tf_mitigation != L1TF_MITIGATION_OFF &&
			e820__mapped_any(half_pa, ULLONG_MAX - half_pa, E820_TYPE_RAM)) {
1121
		pr_warn("System has more than MAX_PA/2 memory. L1TF mitigation not effective.\n");
1122 1123 1124
		pr_info("You may make it effective by booting the kernel with mem=%llu parameter.\n",
				half_pa);
		pr_info("However, doing so will make a part of your RAM unusable.\n");
1125
		pr_info("Reading https://www.kernel.org/doc/html/latest/admin-guide/hw-vuln/l1tf.html might help you decide.\n");
1126 1127 1128 1129 1130
		return;
	}

	setup_force_cpu_cap(X86_FEATURE_L1TF_PTEINV);
}
1131 1132 1133 1134 1135 1136 1137 1138 1139 1140 1141 1142 1143 1144 1145 1146 1147 1148 1149 1150 1151 1152 1153 1154 1155 1156

static int __init l1tf_cmdline(char *str)
{
	if (!boot_cpu_has_bug(X86_BUG_L1TF))
		return 0;

	if (!str)
		return -EINVAL;

	if (!strcmp(str, "off"))
		l1tf_mitigation = L1TF_MITIGATION_OFF;
	else if (!strcmp(str, "flush,nowarn"))
		l1tf_mitigation = L1TF_MITIGATION_FLUSH_NOWARN;
	else if (!strcmp(str, "flush"))
		l1tf_mitigation = L1TF_MITIGATION_FLUSH;
	else if (!strcmp(str, "flush,nosmt"))
		l1tf_mitigation = L1TF_MITIGATION_FLUSH_NOSMT;
	else if (!strcmp(str, "full"))
		l1tf_mitigation = L1TF_MITIGATION_FULL;
	else if (!strcmp(str, "full,force"))
		l1tf_mitigation = L1TF_MITIGATION_FULL_FORCE;

	return 0;
}
early_param("l1tf", l1tf_cmdline);

1157
#undef pr_fmt
1158
#define pr_fmt(fmt) fmt
1159

1160
#ifdef CONFIG_SYSFS
1161

1162 1163 1164
#define L1TF_DEFAULT_MSG "Mitigation: PTE Inversion"

#if IS_ENABLED(CONFIG_KVM_INTEL)
1165
static const char * const l1tf_vmx_states[] = {
1166 1167 1168 1169 1170
	[VMENTER_L1D_FLUSH_AUTO]		= "auto",
	[VMENTER_L1D_FLUSH_NEVER]		= "vulnerable",
	[VMENTER_L1D_FLUSH_COND]		= "conditional cache flushes",
	[VMENTER_L1D_FLUSH_ALWAYS]		= "cache flushes",
	[VMENTER_L1D_FLUSH_EPT_DISABLED]	= "EPT disabled",
1171
	[VMENTER_L1D_FLUSH_NOT_REQUIRED]	= "flush not necessary"
1172 1173 1174 1175 1176 1177 1178
};

static ssize_t l1tf_show_state(char *buf)
{
	if (l1tf_vmx_mitigation == VMENTER_L1D_FLUSH_AUTO)
		return sprintf(buf, "%s\n", L1TF_DEFAULT_MSG);

1179 1180
	if (l1tf_vmx_mitigation == VMENTER_L1D_FLUSH_EPT_DISABLED ||
	    (l1tf_vmx_mitigation == VMENTER_L1D_FLUSH_NEVER &&
T
Thomas Gleixner 已提交
1181
	     sched_smt_active())) {
1182 1183
		return sprintf(buf, "%s; VMX: %s\n", L1TF_DEFAULT_MSG,
			       l1tf_vmx_states[l1tf_vmx_mitigation]);
T
Thomas Gleixner 已提交
1184
	}
1185 1186 1187

	return sprintf(buf, "%s; VMX: %s, SMT %s\n", L1TF_DEFAULT_MSG,
		       l1tf_vmx_states[l1tf_vmx_mitigation],
T
Thomas Gleixner 已提交
1188
		       sched_smt_active() ? "vulnerable" : "disabled");
1189 1190 1191 1192 1193 1194 1195 1196
}
#else
static ssize_t l1tf_show_state(char *buf)
{
	return sprintf(buf, "%s\n", L1TF_DEFAULT_MSG);
}
#endif

1197 1198 1199 1200 1201 1202 1203 1204 1205
static ssize_t mds_show_state(char *buf)
{
	if (!hypervisor_is_type(X86_HYPER_NATIVE)) {
		return sprintf(buf, "%s; SMT Host state unknown\n",
			       mds_strings[mds_mitigation]);
	}

	if (boot_cpu_has(X86_BUG_MSBDS_ONLY)) {
		return sprintf(buf, "%s; SMT %s\n", mds_strings[mds_mitigation],
1206 1207
			       (mds_mitigation == MDS_MITIGATION_OFF ? "vulnerable" :
			        sched_smt_active() ? "mitigated" : "disabled"));
1208 1209 1210 1211 1212 1213
	}

	return sprintf(buf, "%s; SMT %s\n", mds_strings[mds_mitigation],
		       sched_smt_active() ? "vulnerable" : "disabled");
}

1214 1215
static char *stibp_state(void)
{
1216 1217 1218
	if (spectre_v2_enabled == SPECTRE_V2_IBRS_ENHANCED)
		return "";

1219 1220 1221 1222 1223
	switch (spectre_v2_user) {
	case SPECTRE_V2_USER_NONE:
		return ", STIBP: disabled";
	case SPECTRE_V2_USER_STRICT:
		return ", STIBP: forced";
1224
	case SPECTRE_V2_USER_PRCTL:
1225
	case SPECTRE_V2_USER_SECCOMP:
1226 1227
		if (static_key_enabled(&switch_to_cond_stibp))
			return ", STIBP: conditional";
1228 1229
	}
	return "";
1230 1231 1232 1233
}

static char *ibpb_state(void)
{
1234
	if (boot_cpu_has(X86_FEATURE_IBPB)) {
1235
		if (static_key_enabled(&switch_mm_always_ibpb))
1236
			return ", IBPB: always-on";
1237 1238 1239
		if (static_key_enabled(&switch_mm_cond_ibpb))
			return ", IBPB: conditional";
		return ", IBPB: disabled";
1240 1241
	}
	return "";
1242 1243
}

1244
static ssize_t cpu_show_common(struct device *dev, struct device_attribute *attr,
1245
			       char *buf, unsigned int bug)
1246
{
1247
	if (!boot_cpu_has_bug(bug))
1248
		return sprintf(buf, "Not affected\n");
1249 1250 1251 1252 1253 1254

	switch (bug) {
	case X86_BUG_CPU_MELTDOWN:
		if (boot_cpu_has(X86_FEATURE_PTI))
			return sprintf(buf, "Mitigation: PTI\n");

1255 1256 1257
		if (hypervisor_is_type(X86_HYPER_XEN_PV))
			return sprintf(buf, "Unknown (XEN PV detected, hypervisor mitigation required)\n");

1258 1259 1260 1261 1262 1263
		break;

	case X86_BUG_SPECTRE_V1:
		return sprintf(buf, "Mitigation: __user pointer sanitization\n");

	case X86_BUG_SPECTRE_V2:
1264
		return sprintf(buf, "%s%s%s%s%s%s\n", spectre_v2_strings[spectre_v2_enabled],
1265
			       ibpb_state(),
1266
			       boot_cpu_has(X86_FEATURE_USE_IBRS_FW) ? ", IBRS_FW" : "",
1267
			       stibp_state(),
1268
			       boot_cpu_has(X86_FEATURE_RSB_CTXSW) ? ", RSB filling" : "",
1269 1270
			       spectre_v2_module_string());

1271 1272 1273
	case X86_BUG_SPEC_STORE_BYPASS:
		return sprintf(buf, "%s\n", ssb_strings[ssb_mode]);

1274 1275
	case X86_BUG_L1TF:
		if (boot_cpu_has(X86_FEATURE_L1TF_PTEINV))
1276
			return l1tf_show_state(buf);
1277
		break;
1278 1279 1280 1281

	case X86_BUG_MDS:
		return mds_show_state(buf);

1282 1283 1284 1285
	default:
		break;
	}

1286 1287 1288
	return sprintf(buf, "Vulnerable\n");
}

1289 1290 1291 1292 1293
ssize_t cpu_show_meltdown(struct device *dev, struct device_attribute *attr, char *buf)
{
	return cpu_show_common(dev, attr, buf, X86_BUG_CPU_MELTDOWN);
}

1294
ssize_t cpu_show_spectre_v1(struct device *dev, struct device_attribute *attr, char *buf)
1295
{
1296
	return cpu_show_common(dev, attr, buf, X86_BUG_SPECTRE_V1);
1297 1298
}

1299
ssize_t cpu_show_spectre_v2(struct device *dev, struct device_attribute *attr, char *buf)
1300
{
1301
	return cpu_show_common(dev, attr, buf, X86_BUG_SPECTRE_V2);
1302
}
1303 1304 1305 1306 1307

ssize_t cpu_show_spec_store_bypass(struct device *dev, struct device_attribute *attr, char *buf)
{
	return cpu_show_common(dev, attr, buf, X86_BUG_SPEC_STORE_BYPASS);
}
1308 1309 1310 1311 1312

ssize_t cpu_show_l1tf(struct device *dev, struct device_attribute *attr, char *buf)
{
	return cpu_show_common(dev, attr, buf, X86_BUG_L1TF);
}
1313 1314 1315 1316 1317

ssize_t cpu_show_mds(struct device *dev, struct device_attribute *attr, char *buf)
{
	return cpu_show_common(dev, attr, buf, X86_BUG_MDS);
}
1318
#endif