bugs.c 24.9 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

18
#include <asm/spec-ctrl.h>
19
#include <asm/cmdline.h>
20
#include <asm/bugs.h>
21
#include <asm/processor.h>
D
Dave Jones 已提交
22
#include <asm/processor-flags.h>
23
#include <asm/fpu/internal.h>
24
#include <asm/msr.h>
25
#include <asm/vmx.h>
26 27
#include <asm/paravirt.h>
#include <asm/alternative.h>
28
#include <asm/pgtable.h>
L
Laura Abbott 已提交
29
#include <asm/set_memory.h>
30
#include <asm/intel-family.h>
31
#include <asm/e820/api.h>
32
#include <asm/hypervisor.h>
33

34
static void __init spectre_v2_select_mitigation(void);
35
static void __init ssb_select_mitigation(void);
36
static void __init l1tf_select_mitigation(void);
37

38 39 40 41 42
/*
 * Our boot-time value of the SPEC_CTRL MSR. We read it once so that any
 * writes to SPEC_CTRL contain whatever reserved bits have been set.
 */
u64 __ro_after_init x86_spec_ctrl_base;
43
EXPORT_SYMBOL_GPL(x86_spec_ctrl_base);
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 60
void __init check_bugs(void)
{
	identify_boot_cpu();
61

62 63 64 65
	/*
	 * identify_boot_cpu() initialized SMT support information, let the
	 * core code know.
	 */
66
	cpu_smt_check_topology_early();
67

68 69 70 71 72
	if (!IS_ENABLED(CONFIG_SMP)) {
		pr_info("CPU: ");
		print_cpu_info(&boot_cpu_data);
	}

73 74
	/*
	 * Read the SPEC_CTRL MSR to account for reserved bits which may
75 76
	 * 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.
77
	 */
78
	if (boot_cpu_has(X86_FEATURE_MSR_SPEC_CTRL))
79 80
		rdmsrl(MSR_IA32_SPEC_CTRL, x86_spec_ctrl_base);

81 82 83 84
	/* Allow STIBP in MSR_SPEC_CTRL if supported */
	if (boot_cpu_has(X86_FEATURE_STIBP))
		x86_spec_ctrl_mask |= SPEC_CTRL_STIBP;

85 86 87
	/* Select the proper spectre mitigation before patching alternatives */
	spectre_v2_select_mitigation();

88 89 90 91 92 93
	/*
	 * Select proper mitigation for any exposure to the Speculative Store
	 * Bypass vulnerability.
	 */
	ssb_select_mitigation();

94 95
	l1tf_select_mitigation();

96
#ifdef CONFIG_X86_32
97 98 99 100 101 102 103 104 105 106
	/*
	 * 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
107 108
	init_utsname()->machine[1] =
		'0' + (boot_cpu_data.x86 > 6 ? 6 : boot_cpu_data.x86);
109
	alternative_instructions();
110

111
	fpu__init_check_bugs();
112 113 114 115 116 117 118 119 120 121 122 123 124 125
#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
126
}
127

128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143
/* The kernel command line selection */
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,
};

static const char *spectre_v2_strings[] = {
	[SPECTRE_V2_NONE]			= "Vulnerable",
	[SPECTRE_V2_RETPOLINE_MINIMAL]		= "Vulnerable: Minimal generic ASM retpoline",
	[SPECTRE_V2_RETPOLINE_MINIMAL_AMD]	= "Vulnerable: Minimal AMD ASM retpoline",
	[SPECTRE_V2_RETPOLINE_GENERIC]		= "Mitigation: Full generic retpoline",
	[SPECTRE_V2_RETPOLINE_AMD]		= "Mitigation: Full AMD retpoline",
144
	[SPECTRE_V2_IBRS_ENHANCED]		= "Mitigation: Enhanced IBRS",
145 146 147
};

#undef pr_fmt
148
#define pr_fmt(fmt)     "Spectre V2 : " fmt
149

150 151
static enum spectre_v2_mitigation spectre_v2_enabled __ro_after_init =
	SPECTRE_V2_NONE;
152

153 154
void
x86_virt_spec_ctrl(u64 guest_spec_ctrl, u64 guest_virt_spec_ctrl, bool setguest)
155
{
156
	u64 msrval, guestval, hostval = x86_spec_ctrl_base;
157
	struct thread_info *ti = current_thread_info();
158

159
	/* Is MSR_SPEC_CTRL implemented ? */
160
	if (static_cpu_has(X86_FEATURE_MSR_SPEC_CTRL)) {
161 162 163 164 165 166 167 168
		/*
		 * 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;

169
		/* SSBD controlled in MSR_SPEC_CTRL */
170 171
		if (static_cpu_has(X86_FEATURE_SPEC_CTRL_SSBD) ||
		    static_cpu_has(X86_FEATURE_AMD_SSBD))
172
			hostval |= ssbd_tif_to_spec_ctrl(ti->flags);
173

174 175 176
		if (hostval != guestval) {
			msrval = setguest ? guestval : hostval;
			wrmsrl(MSR_IA32_SPEC_CTRL, msrval);
177 178
		}
	}
179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208

	/*
	 * 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);

		speculative_store_bypass_update(tif);
	}
209
}
210
EXPORT_SYMBOL_GPL(x86_virt_spec_ctrl);
211

212
static void x86_amd_ssb_disable(void)
213
{
214
	u64 msrval = x86_amd_ls_cfg_base | x86_amd_ls_cfg_ssbd_mask;
215

216 217 218
	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))
219 220 221
		wrmsrl(MSR_AMD64_LS_CFG, msrval);
}

222
#ifdef RETPOLINE
223 224
static bool spectre_v2_bad_module;

225 226 227 228 229
bool retpoline_module_ok(bool has_retpoline)
{
	if (spectre_v2_enabled == SPECTRE_V2_NONE || has_retpoline)
		return true;

230
	pr_err("System may be vulnerable to spectre v2\n");
231 232 233
	spectre_v2_bad_module = true;
	return false;
}
234 235 236 237 238 239 240

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 ""; }
241
#endif
242 243 244 245

static void __init spec2_print_if_insecure(const char *reason)
{
	if (boot_cpu_has_bug(X86_BUG_SPECTRE_V2))
246
		pr_info("%s selected on command line.\n", reason);
247 248 249 250 251
}

static void __init spec2_print_if_secure(const char *reason)
{
	if (!boot_cpu_has_bug(X86_BUG_SPECTRE_V2))
252
		pr_info("%s selected on command line.\n", reason);
253 254 255 256 257 258 259 260 261 262 263 264 265 266
}

static inline bool retp_compiler(void)
{
	return __is_defined(RETPOLINE);
}

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

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

267 268 269 270 271 272 273 274 275 276 277 278 279
static const struct {
	const char *option;
	enum spectre_v2_mitigation_cmd cmd;
	bool secure;
} mitigation_options[] = {
	{ "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 },
};

280 281 282
static enum spectre_v2_mitigation_cmd __init spectre_v2_parse_cmdline(void)
{
	char arg[20];
283 284 285 286 287 288
	int ret, i;
	enum spectre_v2_mitigation_cmd cmd = SPECTRE_V2_CMD_AUTO;

	if (cmdline_find_option_bool(boot_command_line, "nospectre_v2"))
		return SPECTRE_V2_CMD_NONE;
	else {
289
		ret = cmdline_find_option(boot_command_line, "spectre_v2", arg, sizeof(arg));
290 291 292 293 294 295 296 297 298 299 300
		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)) {
301
			pr_err("unknown option (%s). Switching to AUTO select\n", arg);
302 303 304 305
			return SPECTRE_V2_CMD_AUTO;
		}
	}

306 307 308 309
	if ((cmd == SPECTRE_V2_CMD_RETPOLINE ||
	     cmd == SPECTRE_V2_CMD_RETPOLINE_AMD ||
	     cmd == SPECTRE_V2_CMD_RETPOLINE_GENERIC) &&
	    !IS_ENABLED(CONFIG_RETPOLINE)) {
310
		pr_err("%s selected but not compiled in. Switching to AUTO select\n", mitigation_options[i].option);
311
		return SPECTRE_V2_CMD_AUTO;
312 313 314 315 316 317 318 319 320 321 322 323 324 325
	}

	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;
	}

	if (mitigation_options[i].secure)
		spec2_print_if_secure(mitigation_options[i].option);
	else
		spec2_print_if_insecure(mitigation_options[i].option);

	return cmd;
326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346
}

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:
347 348 349 350 351 352 353
		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;
		}
354 355 356
		if (IS_ENABLED(CONFIG_RETPOLINE))
			goto retpoline_auto;
		break;
357 358 359 360 361 362 363 364 365 366 367 368 369
	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;
	}
370
	pr_err("Spectre mitigation: kernel not compiled with retpoline; no mitigation available!");
371 372 373 374 375 376
	return;

retpoline_auto:
	if (boot_cpu_data.x86_vendor == X86_VENDOR_AMD) {
	retpoline_amd:
		if (!boot_cpu_has(X86_FEATURE_LFENCE_RDTSC)) {
377
			pr_err("Spectre mitigation: LFENCE not serializing, switching to generic retpoline\n");
378 379 380 381 382 383 384 385 386 387 388 389 390
			goto retpoline_generic;
		}
		mode = retp_compiler() ? SPECTRE_V2_RETPOLINE_AMD :
					 SPECTRE_V2_RETPOLINE_MINIMAL_AMD;
		setup_force_cpu_cap(X86_FEATURE_RETPOLINE_AMD);
		setup_force_cpu_cap(X86_FEATURE_RETPOLINE);
	} else {
	retpoline_generic:
		mode = retp_compiler() ? SPECTRE_V2_RETPOLINE_GENERIC :
					 SPECTRE_V2_RETPOLINE_MINIMAL;
		setup_force_cpu_cap(X86_FEATURE_RETPOLINE);
	}

391
specv2_set_mode:
392 393
	spectre_v2_enabled = mode;
	pr_info("%s\n", spectre_v2_strings[mode]);
394 395

	/*
396 397 398
	 * If spectre v2 protection has been enabled, unconditionally fill
	 * RSB during a context switch; this protects against two independent
	 * issues:
399
	 *
400 401
	 *	- RSB underflow (and switch to BTB) on Skylake+
	 *	- SpectreRSB variant of spectre v2 on X86_BUG_SPECTRE_V2 CPUs
402
	 */
403 404
	setup_force_cpu_cap(X86_FEATURE_RSB_CTXSW);
	pr_info("Spectre v2 / SpectreRSB mitigation: Filling RSB on context switch\n");
405 406

	/* Initialize Indirect Branch Prediction Barrier if supported */
407 408
	if (boot_cpu_has(X86_FEATURE_IBPB)) {
		setup_force_cpu_cap(X86_FEATURE_USE_IBPB);
409
		pr_info("Spectre v2 mitigation: Enabling Indirect Branch Prediction Barrier\n");
410
	}
411 412 413

	/*
	 * Retpoline means the kernel is safe because it has no indirect
414 415 416 417 418 419 420 421
	 * 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.
422
	 */
423
	if (boot_cpu_has(X86_FEATURE_IBRS) && mode != SPECTRE_V2_IBRS_ENHANCED) {
424 425 426
		setup_force_cpu_cap(X86_FEATURE_USE_IBRS_FW);
		pr_info("Enabling Restricted Speculation for firmware calls\n");
	}
427 428
}

429 430 431
#undef pr_fmt
#define pr_fmt(fmt)	"Speculative Store Bypass: " fmt

432
static enum ssb_mitigation ssb_mode __ro_after_init = SPEC_STORE_BYPASS_NONE;
433 434 435 436 437 438

/* The kernel command line selection */
enum ssb_mitigation_cmd {
	SPEC_STORE_BYPASS_CMD_NONE,
	SPEC_STORE_BYPASS_CMD_AUTO,
	SPEC_STORE_BYPASS_CMD_ON,
439
	SPEC_STORE_BYPASS_CMD_PRCTL,
440
	SPEC_STORE_BYPASS_CMD_SECCOMP,
441 442 443 444
};

static const char *ssb_strings[] = {
	[SPEC_STORE_BYPASS_NONE]	= "Vulnerable",
445
	[SPEC_STORE_BYPASS_DISABLE]	= "Mitigation: Speculative Store Bypass disabled",
446 447
	[SPEC_STORE_BYPASS_PRCTL]	= "Mitigation: Speculative Store Bypass disabled via prctl",
	[SPEC_STORE_BYPASS_SECCOMP]	= "Mitigation: Speculative Store Bypass disabled via prctl and seccomp",
448 449 450 451 452 453
};

static const struct {
	const char *option;
	enum ssb_mitigation_cmd cmd;
} ssb_mitigation_options[] = {
454 455 456 457 458
	{ "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 */
459 460 461 462 463 464 465 466 467 468 469 470 471 472 473 474 475 476 477 478 479 480 481 482 483 484 485 486 487 488 489 490 491
};

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;

	if (cmdline_find_option_bool(boot_command_line, "nospec_store_bypass_disable")) {
		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;
}

492
static enum ssb_mitigation __init __ssb_select_mitigation(void)
493 494 495 496
{
	enum ssb_mitigation mode = SPEC_STORE_BYPASS_NONE;
	enum ssb_mitigation_cmd cmd;

497
	if (!boot_cpu_has(X86_FEATURE_SSBD))
498 499 500 501 502 503 504 505 506 507
		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:
508 509 510 511 512 513 514 515 516
	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;
517
		break;
518 519 520
	case SPEC_STORE_BYPASS_CMD_ON:
		mode = SPEC_STORE_BYPASS_DISABLE;
		break;
521 522 523
	case SPEC_STORE_BYPASS_CMD_PRCTL:
		mode = SPEC_STORE_BYPASS_PRCTL;
		break;
524 525 526 527
	case SPEC_STORE_BYPASS_CMD_NONE:
		break;
	}

528 529 530
	/*
	 * We have three CPU feature flags that are in play here:
	 *  - X86_BUG_SPEC_STORE_BYPASS - CPU is susceptible.
531
	 *  - X86_FEATURE_SSBD - CPU is able to turn off speculative store bypass
532 533
	 *  - X86_FEATURE_SPEC_STORE_BYPASS_DISABLE - engage the mitigation
	 */
534
	if (mode == SPEC_STORE_BYPASS_DISABLE) {
535
		setup_force_cpu_cap(X86_FEATURE_SPEC_STORE_BYPASS_DISABLE);
536
		/*
537 538
		 * Intel uses the SPEC CTRL MSR Bit(2) for this, while AMD may
		 * use a completely different MSR and bit dependent on family.
539
		 */
540 541
		if (!static_cpu_has(X86_FEATURE_SPEC_CTRL_SSBD) &&
		    !static_cpu_has(X86_FEATURE_AMD_SSBD)) {
542
			x86_amd_ssb_disable();
543
		} else {
544
			x86_spec_ctrl_base |= SPEC_CTRL_SSBD;
545
			x86_spec_ctrl_mask |= SPEC_CTRL_SSBD;
546
			wrmsrl(MSR_IA32_SPEC_CTRL, x86_spec_ctrl_base);
547 548 549
		}
	}

550 551 552
	return mode;
}

553
static void ssb_select_mitigation(void)
554 555 556 557 558 559 560
{
	ssb_mode = __ssb_select_mitigation();

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

561
#undef pr_fmt
562
#define pr_fmt(fmt)     "Speculation prctl: " fmt
563

564
static int ssb_prctl_set(struct task_struct *task, unsigned long ctrl)
565
{
566
	bool update;
567

568 569
	if (ssb_mode != SPEC_STORE_BYPASS_PRCTL &&
	    ssb_mode != SPEC_STORE_BYPASS_SECCOMP)
570 571
		return -ENXIO;

572 573 574 575 576 577
	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);
578
		update = test_and_clear_tsk_thread_flag(task, TIF_SSBD);
579 580 581
		break;
	case PR_SPEC_DISABLE:
		task_set_spec_ssb_disable(task);
582
		update = !test_and_set_tsk_thread_flag(task, TIF_SSBD);
583 584 585 586
		break;
	case PR_SPEC_FORCE_DISABLE:
		task_set_spec_ssb_disable(task);
		task_set_spec_ssb_force_disable(task);
587
		update = !test_and_set_tsk_thread_flag(task, TIF_SSBD);
588 589 590 591
		break;
	default:
		return -ERANGE;
	}
592

593 594 595 596
	/*
	 * If being set on non-current task, delay setting the CPU
	 * mitigation until it is next scheduled.
	 */
597
	if (task == current && update)
598
		speculative_store_bypass_update_current();
599 600 601 602

	return 0;
}

603 604 605 606 607 608 609 610 611 612 613 614 615 616
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);
	default:
		return -ENODEV;
	}
}

#ifdef CONFIG_SECCOMP
void arch_seccomp_spec_mitigate(struct task_struct *task)
{
617 618
	if (ssb_mode == SPEC_STORE_BYPASS_SECCOMP)
		ssb_prctl_set(task, PR_SPEC_FORCE_DISABLE);
619 620 621
}
#endif

622
static int ssb_prctl_get(struct task_struct *task)
623 624 625 626
{
	switch (ssb_mode) {
	case SPEC_STORE_BYPASS_DISABLE:
		return PR_SPEC_DISABLE;
627
	case SPEC_STORE_BYPASS_SECCOMP:
628
	case SPEC_STORE_BYPASS_PRCTL:
629 630 631
		if (task_spec_ssb_force_disable(task))
			return PR_SPEC_PRCTL | PR_SPEC_FORCE_DISABLE;
		if (task_spec_ssb_disable(task))
632 633 634 635 636 637 638 639 640
			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;
	}
}

641
int arch_prctl_spec_ctrl_get(struct task_struct *task, unsigned long which)
642 643 644
{
	switch (which) {
	case PR_SPEC_STORE_BYPASS:
645
		return ssb_prctl_get(task);
646 647 648 649 650
	default:
		return -ENODEV;
	}
}

651 652
void x86_spec_ctrl_setup_ap(void)
{
653
	if (boot_cpu_has(X86_FEATURE_MSR_SPEC_CTRL))
654
		wrmsrl(MSR_IA32_SPEC_CTRL, x86_spec_ctrl_base);
655 656

	if (ssb_mode == SPEC_STORE_BYPASS_DISABLE)
657
		x86_amd_ssb_disable();
658 659
}

660 661
#undef pr_fmt
#define pr_fmt(fmt)	"L1TF: " fmt
662

663 664
/* Default mitigation for L1TF-affected CPUs */
enum l1tf_mitigations l1tf_mitigation __ro_after_init = L1TF_MITIGATION_FLUSH;
665
#if IS_ENABLED(CONFIG_KVM_INTEL)
666
EXPORT_SYMBOL_GPL(l1tf_mitigation);
667
#endif
668
enum vmx_l1d_flush_state l1tf_vmx_mitigation = VMENTER_L1D_FLUSH_AUTO;
669 670
EXPORT_SYMBOL_GPL(l1tf_vmx_mitigation);

671 672 673 674 675 676 677 678 679 680 681 682 683 684 685 686 687 688 689 690 691 692 693 694 695 696 697 698 699 700 701 702 703 704 705 706 707 708 709
/*
 * 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;
	}
}

710 711 712 713 714 715 716
static void __init l1tf_select_mitigation(void)
{
	u64 half_pa;

	if (!boot_cpu_has_bug(X86_BUG_L1TF))
		return;

717 718
	override_cache_bits(&boot_cpu_data);

719 720 721 722 723 724 725 726 727 728 729 730 731 732
	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;
	}

733 734 735 736 737 738 739 740
#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;
	if (e820__mapped_any(half_pa, ULLONG_MAX - half_pa, E820_TYPE_RAM)) {
		pr_warn("System has more than MAX_PA/2 memory. L1TF mitigation not effective.\n");
741 742 743 744
		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");
		pr_info("Reading https://www.kernel.org/doc/html/latest/admin-guide/l1tf.html might help you decide.\n");
745 746 747 748 749
		return;
	}

	setup_force_cpu_cap(X86_FEATURE_L1TF_PTEINV);
}
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

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);

776 777
#undef pr_fmt

778
#ifdef CONFIG_SYSFS
779

780 781 782 783
#define L1TF_DEFAULT_MSG "Mitigation: PTE Inversion"

#if IS_ENABLED(CONFIG_KVM_INTEL)
static const char *l1tf_vmx_states[] = {
784 785 786 787 788
	[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",
789
	[VMENTER_L1D_FLUSH_NOT_REQUIRED]	= "flush not necessary"
790 791 792 793 794 795 796
};

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

797 798 799 800 801 802 803 804 805
	if (l1tf_vmx_mitigation == VMENTER_L1D_FLUSH_EPT_DISABLED ||
	    (l1tf_vmx_mitigation == VMENTER_L1D_FLUSH_NEVER &&
	     cpu_smt_control == CPU_SMT_ENABLED))
		return sprintf(buf, "%s; VMX: %s\n", L1TF_DEFAULT_MSG,
			       l1tf_vmx_states[l1tf_vmx_mitigation]);

	return sprintf(buf, "%s; VMX: %s, SMT %s\n", L1TF_DEFAULT_MSG,
		       l1tf_vmx_states[l1tf_vmx_mitigation],
		       cpu_smt_control == CPU_SMT_ENABLED ? "vulnerable" : "disabled");
806 807 808 809 810 811 812 813
}
#else
static ssize_t l1tf_show_state(char *buf)
{
	return sprintf(buf, "%s\n", L1TF_DEFAULT_MSG);
}
#endif

814
static ssize_t cpu_show_common(struct device *dev, struct device_attribute *attr,
815
			       char *buf, unsigned int bug)
816
{
817
	if (!boot_cpu_has_bug(bug))
818
		return sprintf(buf, "Not affected\n");
819 820 821 822 823 824

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

825 826 827
		if (hypervisor_is_type(X86_HYPER_XEN_PV))
			return sprintf(buf, "Unknown (XEN PV detected, hypervisor mitigation required)\n");

828 829 830 831 832 833
		break;

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

	case X86_BUG_SPECTRE_V2:
834
		return sprintf(buf, "%s%s%s%s\n", spectre_v2_strings[spectre_v2_enabled],
835 836 837 838
			       boot_cpu_has(X86_FEATURE_USE_IBPB) ? ", IBPB" : "",
			       boot_cpu_has(X86_FEATURE_USE_IBRS_FW) ? ", IBRS_FW" : "",
			       spectre_v2_module_string());

839 840 841
	case X86_BUG_SPEC_STORE_BYPASS:
		return sprintf(buf, "%s\n", ssb_strings[ssb_mode]);

842 843
	case X86_BUG_L1TF:
		if (boot_cpu_has(X86_FEATURE_L1TF_PTEINV))
844
			return l1tf_show_state(buf);
845
		break;
846 847 848 849
	default:
		break;
	}

850 851 852
	return sprintf(buf, "Vulnerable\n");
}

853 854 855 856 857
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);
}

858
ssize_t cpu_show_spectre_v1(struct device *dev, struct device_attribute *attr, char *buf)
859
{
860
	return cpu_show_common(dev, attr, buf, X86_BUG_SPECTRE_V1);
861 862
}

863
ssize_t cpu_show_spectre_v2(struct device *dev, struct device_attribute *attr, char *buf)
864
{
865
	return cpu_show_common(dev, attr, buf, X86_BUG_SPECTRE_V2);
866
}
867 868 869 870 871

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);
}
872 873 874 875 876

ssize_t cpu_show_l1tf(struct device *dev, struct device_attribute *attr, char *buf)
{
	return cpu_show_common(dev, attr, buf, X86_BUG_L1TF);
}
877
#endif