aboutsummaryrefslogtreecommitdiff
path: root/arch/x86/kvm/vmx/sgx.c
blob: 35e7ec91ae86414a14fb9f7459f612ccb5c5946a (plain)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
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
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
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
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
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
492
493
494
495
496
// SPDX-License-Identifier: GPL-2.0
/*  Copyright(c) 2021 Intel Corporation. */

#include <asm/sgx.h>

#include "cpuid.h"
#include "kvm_cache_regs.h"
#include "nested.h"
#include "sgx.h"
#include "vmx.h"
#include "x86.h"

bool __read_mostly enable_sgx = 1;
module_param_named(sgx, enable_sgx, bool, 0444);

/* Initial value of guest's virtual SGX_LEPUBKEYHASHn MSRs */
static u64 sgx_pubkey_hash[4] __ro_after_init;

/*
 * ENCLS's memory operands use a fixed segment (DS) and a fixed
 * address size based on the mode.  Related prefixes are ignored.
 */
static int sgx_get_encls_gva(struct kvm_vcpu *vcpu, unsigned long offset,
			     int size, int alignment, gva_t *gva)
{
	struct kvm_segment s;
	bool fault;

	/* Skip vmcs.GUEST_DS retrieval for 64-bit mode to avoid VMREADs. */
	*gva = offset;
	if (!is_long_mode(vcpu)) {
		vmx_get_segment(vcpu, &s, VCPU_SREG_DS);
		*gva += s.base;
	}

	if (!IS_ALIGNED(*gva, alignment)) {
		fault = true;
	} else if (likely(is_long_mode(vcpu))) {
		fault = is_noncanonical_address(*gva, vcpu);
	} else {
		*gva &= 0xffffffff;
		fault = (s.unusable) ||
			(s.type != 2 && s.type != 3) ||
			(*gva > s.limit) ||
			((s.base != 0 || s.limit != 0xffffffff) &&
			(((u64)*gva + size - 1) > s.limit + 1));
	}
	if (fault)
		kvm_inject_gp(vcpu, 0);
	return fault ? -EINVAL : 0;
}

static void sgx_handle_emulation_failure(struct kvm_vcpu *vcpu, u64 addr,
					 unsigned int size)
{
	uint64_t data[2] = { addr, size };

	__kvm_prepare_emulation_failure_exit(vcpu, data, ARRAY_SIZE(data));
}

static int sgx_read_hva(struct kvm_vcpu *vcpu, unsigned long hva, void *data,
			unsigned int size)
{
	if (__copy_from_user(data, (void __user *)hva, size)) {
		sgx_handle_emulation_failure(vcpu, hva, size);
		return -EFAULT;
	}

	return 0;
}

static int sgx_gva_to_gpa(struct kvm_vcpu *vcpu, gva_t gva, bool write,
			  gpa_t *gpa)
{
	struct x86_exception ex;

	if (write)
		*gpa = kvm_mmu_gva_to_gpa_write(vcpu, gva, &ex);
	else
		*gpa = kvm_mmu_gva_to_gpa_read(vcpu, gva, &ex);

	if (*gpa == UNMAPPED_GVA) {
		kvm_inject_emulated_page_fault(vcpu, &ex);
		return -EFAULT;
	}

	return 0;
}

static int sgx_gpa_to_hva(struct kvm_vcpu *vcpu, gpa_t gpa, unsigned long *hva)
{
	*hva = kvm_vcpu_gfn_to_hva(vcpu, PFN_DOWN(gpa));
	if (kvm_is_error_hva(*hva)) {
		sgx_handle_emulation_failure(vcpu, gpa, 1);
		return -EFAULT;
	}

	*hva |= gpa & ~PAGE_MASK;

	return 0;
}

static int sgx_inject_fault(struct kvm_vcpu *vcpu, gva_t gva, int trapnr)
{
	struct x86_exception ex;

	/*
	 * A non-EPCM #PF indicates a bad userspace HVA.  This *should* check
	 * for PFEC.SGX and not assume any #PF on SGX2 originated in the EPC,
	 * but the error code isn't (yet) plumbed through the ENCLS helpers.
	 */
	if (trapnr == PF_VECTOR && !boot_cpu_has(X86_FEATURE_SGX2)) {
		kvm_prepare_emulation_failure_exit(vcpu);
		return 0;
	}

	/*
	 * If the guest thinks it's running on SGX2 hardware, inject an SGX
	 * #PF if the fault matches an EPCM fault signature (#GP on SGX1,
	 * #PF on SGX2).  The assumption is that EPCM faults are much more
	 * likely than a bad userspace address.
	 */
	if ((trapnr == PF_VECTOR || !boot_cpu_has(X86_FEATURE_SGX2)) &&
	    guest_cpuid_has(vcpu, X86_FEATURE_SGX2)) {
		memset(&ex, 0, sizeof(ex));
		ex.vector = PF_VECTOR;
		ex.error_code = PFERR_PRESENT_MASK | PFERR_WRITE_MASK |
				PFERR_SGX_MASK;
		ex.address = gva;
		ex.error_code_valid = true;
		ex.nested_page_fault = false;
		kvm_inject_page_fault(vcpu, &ex);
	} else {
		kvm_inject_gp(vcpu, 0);
	}
	return 1;
}

static int __handle_encls_ecreate(struct kvm_vcpu *vcpu,
				  struct sgx_pageinfo *pageinfo,
				  unsigned long secs_hva,
				  gva_t secs_gva)
{
	struct sgx_secs *contents = (struct sgx_secs *)pageinfo->contents;
	struct kvm_cpuid_entry2 *sgx_12_0, *sgx_12_1;
	u64 attributes, xfrm, size;
	u32 miscselect;
	u8 max_size_log2;
	int trapnr, ret;

	sgx_12_0 = kvm_find_cpuid_entry(vcpu, 0x12, 0);
	sgx_12_1 = kvm_find_cpuid_entry(vcpu, 0x12, 1);
	if (!sgx_12_0 || !sgx_12_1) {
		kvm_prepare_emulation_failure_exit(vcpu);
		return 0;
	}

	miscselect = contents->miscselect;
	attributes = contents->attributes;
	xfrm = contents->xfrm;
	size = contents->size;

	/* Enforce restriction of access to the PROVISIONKEY. */
	if (!vcpu->kvm->arch.sgx_provisioning_allowed &&
	    (attributes & SGX_ATTR_PROVISIONKEY)) {
		if (sgx_12_1->eax & SGX_ATTR_PROVISIONKEY)
			pr_warn_once("KVM: SGX PROVISIONKEY advertised but not allowed\n");
		kvm_inject_gp(vcpu, 0);
		return 1;
	}

	/* Enforce CPUID restrictions on MISCSELECT, ATTRIBUTES and XFRM. */
	if ((u32)miscselect & ~sgx_12_0->ebx ||
	    (u32)attributes & ~sgx_12_1->eax ||
	    (u32)(attributes >> 32) & ~sgx_12_1->ebx ||
	    (u32)xfrm & ~sgx_12_1->ecx ||
	    (u32)(xfrm >> 32) & ~sgx_12_1->edx) {
		kvm_inject_gp(vcpu, 0);
		return 1;
	}

	/* Enforce CPUID restriction on max enclave size. */
	max_size_log2 = (attributes & SGX_ATTR_MODE64BIT) ? sgx_12_0->edx >> 8 :
							    sgx_12_0->edx;
	if (size >= BIT_ULL(max_size_log2))
		kvm_inject_gp(vcpu, 0);

	/*
	 * sgx_virt_ecreate() returns:
	 *  1) 0:	ECREATE was successful
	 *  2) -EFAULT:	ECREATE was run but faulted, and trapnr was set to the
	 *		exception number.
	 *  3) -EINVAL:	access_ok() on @secs_hva failed. This should never
	 *		happen as KVM checks host addresses at memslot creation.
	 *		sgx_virt_ecreate() has already warned in this case.
	 */
	ret = sgx_virt_ecreate(pageinfo, (void __user *)secs_hva, &trapnr);
	if (!ret)
		return kvm_skip_emulated_instruction(vcpu);
	if (ret == -EFAULT)
		return sgx_inject_fault(vcpu, secs_gva, trapnr);

	return ret;
}

static int handle_encls_ecreate(struct kvm_vcpu *vcpu)
{
	gva_t pageinfo_gva, secs_gva;
	gva_t metadata_gva, contents_gva;
	gpa_t metadata_gpa, contents_gpa, secs_gpa;
	unsigned long metadata_hva, contents_hva, secs_hva;
	struct sgx_pageinfo pageinfo;
	struct sgx_secs *contents;
	struct x86_exception ex;
	int r;

	if (sgx_get_encls_gva(vcpu, kvm_rbx_read(vcpu), 32, 32, &pageinfo_gva) ||
	    sgx_get_encls_gva(vcpu, kvm_rcx_read(vcpu), 4096, 4096, &secs_gva))
		return 1;

	/*
	 * Copy the PAGEINFO to local memory, its pointers need to be
	 * translated, i.e. we need to do a deep copy/translate.
	 */
	r = kvm_read_guest_virt(vcpu, pageinfo_gva, &pageinfo,
				sizeof(pageinfo), &ex);
	if (r == X86EMUL_PROPAGATE_FAULT) {
		kvm_inject_emulated_page_fault(vcpu, &ex);
		return 1;
	} else if (r != X86EMUL_CONTINUE) {
		sgx_handle_emulation_failure(vcpu, pageinfo_gva,
					     sizeof(pageinfo));
		return 0;
	}

	if (sgx_get_encls_gva(vcpu, pageinfo.metadata, 64, 64, &metadata_gva) ||
	    sgx_get_encls_gva(vcpu, pageinfo.contents, 4096, 4096,
			      &contents_gva))
		return 1;

	/*
	 * Translate the SECINFO, SOURCE and SECS pointers from GVA to GPA.
	 * Resume the guest on failure to inject a #PF.
	 */
	if (sgx_gva_to_gpa(vcpu, metadata_gva, false, &metadata_gpa) ||
	    sgx_gva_to_gpa(vcpu, contents_gva, false, &contents_gpa) ||
	    sgx_gva_to_gpa(vcpu, secs_gva, true, &secs_gpa))
		return 1;

	/*
	 * ...and then to HVA.  The order of accesses isn't architectural, i.e.
	 * KVM doesn't have to fully process one address at a time.  Exit to
	 * userspace if a GPA is invalid.
	 */
	if (sgx_gpa_to_hva(vcpu, metadata_gpa, &metadata_hva) ||
	    sgx_gpa_to_hva(vcpu, contents_gpa, &contents_hva) ||
	    sgx_gpa_to_hva(vcpu, secs_gpa, &secs_hva))
		return 0;

	/*
	 * Copy contents into kernel memory to prevent TOCTOU attack. E.g. the
	 * guest could do ECREATE w/ SECS.SGX_ATTR_PROVISIONKEY=0, and
	 * simultaneously set SGX_ATTR_PROVISIONKEY to bypass the check to
	 * enforce restriction of access to the PROVISIONKEY.
	 */
	contents = (struct sgx_secs *)__get_free_page(GFP_KERNEL_ACCOUNT);
	if (!contents)
		return -ENOMEM;

	/* Exit to userspace if copying from a host userspace address fails. */
	if (sgx_read_hva(vcpu, contents_hva, (void *)contents, PAGE_SIZE)) {
		free_page((unsigned long)contents);
		return 0;
	}

	pageinfo.metadata = metadata_hva;
	pageinfo.contents = (u64)contents;

	r = __handle_encls_ecreate(vcpu, &pageinfo, secs_hva, secs_gva);

	free_page((unsigned long)contents);

	return r;
}

static int handle_encls_einit(struct kvm_vcpu *vcpu)
{
	unsigned long sig_hva, secs_hva, token_hva, rflags;
	struct vcpu_vmx *vmx = to_vmx(vcpu);
	gva_t sig_gva, secs_gva, token_gva;
	gpa_t sig_gpa, secs_gpa, token_gpa;
	int ret, trapnr;

	if (sgx_get_encls_gva(vcpu, kvm_rbx_read(vcpu), 1808, 4096, &sig_gva) ||
	    sgx_get_encls_gva(vcpu, kvm_rcx_read(vcpu), 4096, 4096, &secs_gva) ||
	    sgx_get_encls_gva(vcpu, kvm_rdx_read(vcpu), 304, 512, &token_gva))
		return 1;

	/*
	 * Translate the SIGSTRUCT, SECS and TOKEN pointers from GVA to GPA.
	 * Resume the guest on failure to inject a #PF.
	 */
	if (sgx_gva_to_gpa(vcpu, sig_gva, false, &sig_gpa) ||
	    sgx_gva_to_gpa(vcpu, secs_gva, true, &secs_gpa) ||
	    sgx_gva_to_gpa(vcpu, token_gva, false, &token_gpa))
		return 1;

	/*
	 * ...and then to HVA.  The order of accesses isn't architectural, i.e.
	 * KVM doesn't have to fully process one address at a time.  Exit to
	 * userspace if a GPA is invalid.  Note, all structures are aligned and
	 * cannot split pages.
	 */
	if (sgx_gpa_to_hva(vcpu, sig_gpa, &sig_hva) ||
	    sgx_gpa_to_hva(vcpu, secs_gpa, &secs_hva) ||
	    sgx_gpa_to_hva(vcpu, token_gpa, &token_hva))
		return 0;

	ret = sgx_virt_einit((void __user *)sig_hva, (void __user *)token_hva,
			     (void __user *)secs_hva,
			     vmx->msr_ia32_sgxlepubkeyhash, &trapnr);

	if (ret == -EFAULT)
		return sgx_inject_fault(vcpu, secs_gva, trapnr);

	/*
	 * sgx_virt_einit() returns -EINVAL when access_ok() fails on @sig_hva,
	 * @token_hva or @secs_hva. This should never happen as KVM checks host
	 * addresses at memslot creation. sgx_virt_einit() has already warned
	 * in this case, so just return.
	 */
	if (ret < 0)
		return ret;

	rflags = vmx_get_rflags(vcpu) & ~(X86_EFLAGS_CF | X86_EFLAGS_PF |
					  X86_EFLAGS_AF | X86_EFLAGS_SF |
					  X86_EFLAGS_OF);
	if (ret)
		rflags |= X86_EFLAGS_ZF;
	else
		rflags &= ~X86_EFLAGS_ZF;
	vmx_set_rflags(vcpu, rflags);

	kvm_rax_write(vcpu, ret);
	return kvm_skip_emulated_instruction(vcpu);
}

static inline bool encls_leaf_enabled_in_guest(struct kvm_vcpu *vcpu, u32 leaf)
{
	if (!enable_sgx || !guest_cpuid_has(vcpu, X86_FEATURE_SGX))
		return false;

	if (leaf >= ECREATE && leaf <= ETRACK)
		return guest_cpuid_has(vcpu, X86_FEATURE_SGX1);

	if (leaf >= EAUG && leaf <= EMODT)
		return guest_cpuid_has(vcpu, X86_FEATURE_SGX2);

	return false;
}

static inline bool sgx_enabled_in_guest_bios(struct kvm_vcpu *vcpu)
{
	const u64 bits = FEAT_CTL_SGX_ENABLED | FEAT_CTL_LOCKED;

	return (to_vmx(vcpu)->msr_ia32_feature_control & bits) == bits;
}

int handle_encls(struct kvm_vcpu *vcpu)
{
	u32 leaf = (u32)kvm_rax_read(vcpu);

	if (!encls_leaf_enabled_in_guest(vcpu, leaf)) {
		kvm_queue_exception(vcpu, UD_VECTOR);
	} else if (!sgx_enabled_in_guest_bios(vcpu)) {
		kvm_inject_gp(vcpu, 0);
	} else {
		if (leaf == ECREATE)
			return handle_encls_ecreate(vcpu);
		if (leaf == EINIT)
			return handle_encls_einit(vcpu);
		WARN(1, "KVM: unexpected exit on ENCLS[%u]", leaf);
		vcpu->run->exit_reason = KVM_EXIT_UNKNOWN;
		vcpu->run->hw.hardware_exit_reason = EXIT_REASON_ENCLS;
		return 0;
	}
	return 1;
}

void setup_default_sgx_lepubkeyhash(void)
{
	/*
	 * Use Intel's default value for Skylake hardware if Launch Control is
	 * not supported, i.e. Intel's hash is hardcoded into silicon, or if
	 * Launch Control is supported and enabled, i.e. mimic the reset value
	 * and let the guest write the MSRs at will.  If Launch Control is
	 * supported but disabled, then use the current MSR values as the hash
	 * MSRs exist but are read-only (locked and not writable).
	 */
	if (!enable_sgx || boot_cpu_has(X86_FEATURE_SGX_LC) ||
	    rdmsrl_safe(MSR_IA32_SGXLEPUBKEYHASH0, &sgx_pubkey_hash[0])) {
		sgx_pubkey_hash[0] = 0xa6053e051270b7acULL;
		sgx_pubkey_hash[1] = 0x6cfbe8ba8b3b413dULL;
		sgx_pubkey_hash[2] = 0xc4916d99f2b3735dULL;
		sgx_pubkey_hash[3] = 0xd4f8c05909f9bb3bULL;
	} else {
		/* MSR_IA32_SGXLEPUBKEYHASH0 is read above */
		rdmsrl(MSR_IA32_SGXLEPUBKEYHASH1, sgx_pubkey_hash[1]);
		rdmsrl(MSR_IA32_SGXLEPUBKEYHASH2, sgx_pubkey_hash[2]);
		rdmsrl(MSR_IA32_SGXLEPUBKEYHASH3, sgx_pubkey_hash[3]);
	}
}

void vcpu_setup_sgx_lepubkeyhash(struct kvm_vcpu *vcpu)
{
	struct vcpu_vmx *vmx = to_vmx(vcpu);

	memcpy(vmx->msr_ia32_sgxlepubkeyhash, sgx_pubkey_hash,
	       sizeof(sgx_pubkey_hash));
}

/*
 * ECREATE must be intercepted to enforce MISCSELECT, ATTRIBUTES and XFRM
 * restrictions if the guest's allowed-1 settings diverge from hardware.
 */
static bool sgx_intercept_encls_ecreate(struct kvm_vcpu *vcpu)
{
	struct kvm_cpuid_entry2 *guest_cpuid;
	u32 eax, ebx, ecx, edx;

	if (!vcpu->kvm->arch.sgx_provisioning_allowed)
		return true;

	guest_cpuid = kvm_find_cpuid_entry(vcpu, 0x12, 0);
	if (!guest_cpuid)
		return true;

	cpuid_count(0x12, 0, &eax, &ebx, &ecx, &edx);
	if (guest_cpuid->ebx != ebx || guest_cpuid->edx != edx)
		return true;

	guest_cpuid = kvm_find_cpuid_entry(vcpu, 0x12, 1);
	if (!guest_cpuid)
		return true;

	cpuid_count(0x12, 1, &eax, &ebx, &ecx, &edx);
	if (guest_cpuid->eax != eax || guest_cpuid->ebx != ebx ||
	    guest_cpuid->ecx != ecx || guest_cpuid->edx != edx)
		return true;

	return false;
}

void vmx_write_encls_bitmap(struct kvm_vcpu *vcpu, struct vmcs12 *vmcs12)
{
	/*
	 * There is no software enable bit for SGX that is virtualized by
	 * hardware, e.g. there's no CR4.SGXE, so when SGX is disabled in the
	 * guest (either by the host or by the guest's BIOS) but enabled in the
	 * host, trap all ENCLS leafs and inject #UD/#GP as needed to emulate
	 * the expected system behavior for ENCLS.
	 */
	u64 bitmap = -1ull;

	/* Nothing to do if hardware doesn't support SGX */
	if (!cpu_has_vmx_encls_vmexit())
		return;

	if (guest_cpuid_has(vcpu, X86_FEATURE_SGX) &&
	    sgx_enabled_in_guest_bios(vcpu)) {
		if (guest_cpuid_has(vcpu, X86_FEATURE_SGX1)) {
			bitmap &= ~GENMASK_ULL(ETRACK, ECREATE);
			if (sgx_intercept_encls_ecreate(vcpu))
				bitmap |= (1 << ECREATE);
		}

		if (guest_cpuid_has(vcpu, X86_FEATURE_SGX2))
			bitmap &= ~GENMASK_ULL(EMODT, EAUG);

		/*
		 * Trap and execute EINIT if launch control is enabled in the
		 * host using the guest's values for launch control MSRs, even
		 * if the guest's values are fixed to hardware default values.
		 * The MSRs are not loaded/saved on VM-Enter/VM-Exit as writing
		 * the MSRs is extraordinarily expensive.
		 */
		if (boot_cpu_has(X86_FEATURE_SGX_LC))
			bitmap |= (1 << EINIT);

		if (!vmcs12 && is_guest_mode(vcpu))
			vmcs12 = get_vmcs12(vcpu);
		if (vmcs12 && nested_cpu_has_encls_exit(vmcs12))
			bitmap |= vmcs12->encls_exiting_bitmap;
	}
	vmcs_write64(ENCLS_EXITING_BITMAP, bitmap);
}