xref: /qemu/target/i386/machine.c (revision d051d0e1)
1 #include "qemu/osdep.h"
2 #include "cpu.h"
3 #include "exec/exec-all.h"
4 #include "hw/isa/isa.h"
5 #include "migration/cpu.h"
6 #include "kvm/hyperv.h"
7 #include "hw/i386/x86.h"
8 #include "kvm/kvm_i386.h"
9 
10 #include "sysemu/kvm.h"
11 #include "sysemu/tcg.h"
12 
13 #include "qemu/error-report.h"
14 
15 static const VMStateDescription vmstate_segment = {
16     .name = "segment",
17     .version_id = 1,
18     .minimum_version_id = 1,
19     .fields = (VMStateField[]) {
20         VMSTATE_UINT32(selector, SegmentCache),
21         VMSTATE_UINTTL(base, SegmentCache),
22         VMSTATE_UINT32(limit, SegmentCache),
23         VMSTATE_UINT32(flags, SegmentCache),
24         VMSTATE_END_OF_LIST()
25     }
26 };
27 
28 #define VMSTATE_SEGMENT(_field, _state) {                            \
29     .name       = (stringify(_field)),                               \
30     .size       = sizeof(SegmentCache),                              \
31     .vmsd       = &vmstate_segment,                                  \
32     .flags      = VMS_STRUCT,                                        \
33     .offset     = offsetof(_state, _field)                           \
34             + type_check(SegmentCache,typeof_field(_state, _field))  \
35 }
36 
37 #define VMSTATE_SEGMENT_ARRAY(_field, _state, _n)                    \
38     VMSTATE_STRUCT_ARRAY(_field, _state, _n, 0, vmstate_segment, SegmentCache)
39 
40 static const VMStateDescription vmstate_xmm_reg = {
41     .name = "xmm_reg",
42     .version_id = 1,
43     .minimum_version_id = 1,
44     .fields = (VMStateField[]) {
45         VMSTATE_UINT64(ZMM_Q(0), ZMMReg),
46         VMSTATE_UINT64(ZMM_Q(1), ZMMReg),
47         VMSTATE_END_OF_LIST()
48     }
49 };
50 
51 #define VMSTATE_XMM_REGS(_field, _state, _start)                         \
52     VMSTATE_STRUCT_SUB_ARRAY(_field, _state, _start, CPU_NB_REGS, 0,     \
53                              vmstate_xmm_reg, ZMMReg)
54 
55 /* YMMH format is the same as XMM, but for bits 128-255 */
56 static const VMStateDescription vmstate_ymmh_reg = {
57     .name = "ymmh_reg",
58     .version_id = 1,
59     .minimum_version_id = 1,
60     .fields = (VMStateField[]) {
61         VMSTATE_UINT64(ZMM_Q(2), ZMMReg),
62         VMSTATE_UINT64(ZMM_Q(3), ZMMReg),
63         VMSTATE_END_OF_LIST()
64     }
65 };
66 
67 #define VMSTATE_YMMH_REGS_VARS(_field, _state, _start, _v)               \
68     VMSTATE_STRUCT_SUB_ARRAY(_field, _state, _start, CPU_NB_REGS, _v,    \
69                              vmstate_ymmh_reg, ZMMReg)
70 
71 static const VMStateDescription vmstate_zmmh_reg = {
72     .name = "zmmh_reg",
73     .version_id = 1,
74     .minimum_version_id = 1,
75     .fields = (VMStateField[]) {
76         VMSTATE_UINT64(ZMM_Q(4), ZMMReg),
77         VMSTATE_UINT64(ZMM_Q(5), ZMMReg),
78         VMSTATE_UINT64(ZMM_Q(6), ZMMReg),
79         VMSTATE_UINT64(ZMM_Q(7), ZMMReg),
80         VMSTATE_END_OF_LIST()
81     }
82 };
83 
84 #define VMSTATE_ZMMH_REGS_VARS(_field, _state, _start)                   \
85     VMSTATE_STRUCT_SUB_ARRAY(_field, _state, _start, CPU_NB_REGS, 0,     \
86                              vmstate_zmmh_reg, ZMMReg)
87 
88 #ifdef TARGET_X86_64
89 static const VMStateDescription vmstate_hi16_zmm_reg = {
90     .name = "hi16_zmm_reg",
91     .version_id = 1,
92     .minimum_version_id = 1,
93     .fields = (VMStateField[]) {
94         VMSTATE_UINT64(ZMM_Q(0), ZMMReg),
95         VMSTATE_UINT64(ZMM_Q(1), ZMMReg),
96         VMSTATE_UINT64(ZMM_Q(2), ZMMReg),
97         VMSTATE_UINT64(ZMM_Q(3), ZMMReg),
98         VMSTATE_UINT64(ZMM_Q(4), ZMMReg),
99         VMSTATE_UINT64(ZMM_Q(5), ZMMReg),
100         VMSTATE_UINT64(ZMM_Q(6), ZMMReg),
101         VMSTATE_UINT64(ZMM_Q(7), ZMMReg),
102         VMSTATE_END_OF_LIST()
103     }
104 };
105 
106 #define VMSTATE_Hi16_ZMM_REGS_VARS(_field, _state, _start)               \
107     VMSTATE_STRUCT_SUB_ARRAY(_field, _state, _start, CPU_NB_REGS, 0,     \
108                              vmstate_hi16_zmm_reg, ZMMReg)
109 #endif
110 
111 static const VMStateDescription vmstate_bnd_regs = {
112     .name = "bnd_regs",
113     .version_id = 1,
114     .minimum_version_id = 1,
115     .fields = (VMStateField[]) {
116         VMSTATE_UINT64(lb, BNDReg),
117         VMSTATE_UINT64(ub, BNDReg),
118         VMSTATE_END_OF_LIST()
119     }
120 };
121 
122 #define VMSTATE_BND_REGS(_field, _state, _n)          \
123     VMSTATE_STRUCT_ARRAY(_field, _state, _n, 0, vmstate_bnd_regs, BNDReg)
124 
125 static const VMStateDescription vmstate_mtrr_var = {
126     .name = "mtrr_var",
127     .version_id = 1,
128     .minimum_version_id = 1,
129     .fields = (VMStateField[]) {
130         VMSTATE_UINT64(base, MTRRVar),
131         VMSTATE_UINT64(mask, MTRRVar),
132         VMSTATE_END_OF_LIST()
133     }
134 };
135 
136 #define VMSTATE_MTRR_VARS(_field, _state, _n, _v)                    \
137     VMSTATE_STRUCT_ARRAY(_field, _state, _n, _v, vmstate_mtrr_var, MTRRVar)
138 
139 typedef struct x86_FPReg_tmp {
140     FPReg *parent;
141     uint64_t tmp_mant;
142     uint16_t tmp_exp;
143 } x86_FPReg_tmp;
144 
145 static void cpu_get_fp80(uint64_t *pmant, uint16_t *pexp, floatx80 f)
146 {
147     CPU_LDoubleU temp;
148 
149     temp.d = f;
150     *pmant = temp.l.lower;
151     *pexp = temp.l.upper;
152 }
153 
154 static floatx80 cpu_set_fp80(uint64_t mant, uint16_t upper)
155 {
156     CPU_LDoubleU temp;
157 
158     temp.l.upper = upper;
159     temp.l.lower = mant;
160     return temp.d;
161 }
162 
163 static int fpreg_pre_save(void *opaque)
164 {
165     x86_FPReg_tmp *tmp = opaque;
166 
167     /* we save the real CPU data (in case of MMX usage only 'mant'
168        contains the MMX register */
169     cpu_get_fp80(&tmp->tmp_mant, &tmp->tmp_exp, tmp->parent->d);
170 
171     return 0;
172 }
173 
174 static int fpreg_post_load(void *opaque, int version)
175 {
176     x86_FPReg_tmp *tmp = opaque;
177 
178     tmp->parent->d = cpu_set_fp80(tmp->tmp_mant, tmp->tmp_exp);
179     return 0;
180 }
181 
182 static const VMStateDescription vmstate_fpreg_tmp = {
183     .name = "fpreg_tmp",
184     .post_load = fpreg_post_load,
185     .pre_save  = fpreg_pre_save,
186     .fields = (VMStateField[]) {
187         VMSTATE_UINT64(tmp_mant, x86_FPReg_tmp),
188         VMSTATE_UINT16(tmp_exp, x86_FPReg_tmp),
189         VMSTATE_END_OF_LIST()
190     }
191 };
192 
193 static const VMStateDescription vmstate_fpreg = {
194     .name = "fpreg",
195     .fields = (VMStateField[]) {
196         VMSTATE_WITH_TMP(FPReg, x86_FPReg_tmp, vmstate_fpreg_tmp),
197         VMSTATE_END_OF_LIST()
198     }
199 };
200 
201 static int cpu_pre_save(void *opaque)
202 {
203     X86CPU *cpu = opaque;
204     CPUX86State *env = &cpu->env;
205     int i;
206     env->v_tpr = env->int_ctl & V_TPR_MASK;
207     /* FPU */
208     env->fpus_vmstate = (env->fpus & ~0x3800) | (env->fpstt & 0x7) << 11;
209     env->fptag_vmstate = 0;
210     for(i = 0; i < 8; i++) {
211         env->fptag_vmstate |= ((!env->fptags[i]) << i);
212     }
213 
214     env->fpregs_format_vmstate = 0;
215 
216     /*
217      * Real mode guest segments register DPL should be zero.
218      * Older KVM version were setting it wrongly.
219      * Fixing it will allow live migration to host with unrestricted guest
220      * support (otherwise the migration will fail with invalid guest state
221      * error).
222      */
223     if (!(env->cr[0] & CR0_PE_MASK) &&
224         (env->segs[R_CS].flags >> DESC_DPL_SHIFT & 3) != 0) {
225         env->segs[R_CS].flags &= ~(env->segs[R_CS].flags & DESC_DPL_MASK);
226         env->segs[R_DS].flags &= ~(env->segs[R_DS].flags & DESC_DPL_MASK);
227         env->segs[R_ES].flags &= ~(env->segs[R_ES].flags & DESC_DPL_MASK);
228         env->segs[R_FS].flags &= ~(env->segs[R_FS].flags & DESC_DPL_MASK);
229         env->segs[R_GS].flags &= ~(env->segs[R_GS].flags & DESC_DPL_MASK);
230         env->segs[R_SS].flags &= ~(env->segs[R_SS].flags & DESC_DPL_MASK);
231     }
232 
233 #ifdef CONFIG_KVM
234     /*
235      * In case vCPU may have enabled VMX, we need to make sure kernel have
236      * required capabilities in order to perform migration correctly:
237      *
238      * 1) We must be able to extract vCPU nested-state from KVM.
239      *
240      * 2) In case vCPU is running in guest-mode and it has a pending exception,
241      * we must be able to determine if it's in a pending or injected state.
242      * Note that in case KVM don't have required capability to do so,
243      * a pending/injected exception will always appear as an
244      * injected exception.
245      */
246     if (kvm_enabled() && cpu_vmx_maybe_enabled(env) &&
247         (!env->nested_state ||
248          (!kvm_has_exception_payload() && (env->hflags & HF_GUEST_MASK) &&
249           env->exception_injected))) {
250         error_report("Guest maybe enabled nested virtualization but kernel "
251                 "does not support required capabilities to save vCPU "
252                 "nested state");
253         return -EINVAL;
254     }
255 #endif
256 
257     /*
258      * When vCPU is running L2 and exception is still pending,
259      * it can potentially be intercepted by L1 hypervisor.
260      * In contrast to an injected exception which cannot be
261      * intercepted anymore.
262      *
263      * Furthermore, when a L2 exception is intercepted by L1
264      * hypervisor, its exception payload (CR2/DR6 on #PF/#DB)
265      * should not be set yet in the respective vCPU register.
266      * Thus, in case an exception is pending, it is
267      * important to save the exception payload seperately.
268      *
269      * Therefore, if an exception is not in a pending state
270      * or vCPU is not in guest-mode, it is not important to
271      * distinguish between a pending and injected exception
272      * and we don't need to store seperately the exception payload.
273      *
274      * In order to preserve better backwards-compatible migration,
275      * convert a pending exception to an injected exception in
276      * case it is not important to distinguish between them
277      * as described above.
278      */
279     if (env->exception_pending && !(env->hflags & HF_GUEST_MASK)) {
280         env->exception_pending = 0;
281         env->exception_injected = 1;
282 
283         if (env->exception_has_payload) {
284             if (env->exception_nr == EXCP01_DB) {
285                 env->dr[6] = env->exception_payload;
286             } else if (env->exception_nr == EXCP0E_PAGE) {
287                 env->cr[2] = env->exception_payload;
288             }
289         }
290     }
291 
292     return 0;
293 }
294 
295 static int cpu_post_load(void *opaque, int version_id)
296 {
297     X86CPU *cpu = opaque;
298     CPUState *cs = CPU(cpu);
299     CPUX86State *env = &cpu->env;
300     int i;
301 
302     if (env->tsc_khz && env->user_tsc_khz &&
303         env->tsc_khz != env->user_tsc_khz) {
304         error_report("Mismatch between user-specified TSC frequency and "
305                      "migrated TSC frequency");
306         return -EINVAL;
307     }
308 
309     if (env->fpregs_format_vmstate) {
310         error_report("Unsupported old non-softfloat CPU state");
311         return -EINVAL;
312     }
313     /*
314      * Real mode guest segments register DPL should be zero.
315      * Older KVM version were setting it wrongly.
316      * Fixing it will allow live migration from such host that don't have
317      * restricted guest support to a host with unrestricted guest support
318      * (otherwise the migration will fail with invalid guest state
319      * error).
320      */
321     if (!(env->cr[0] & CR0_PE_MASK) &&
322         (env->segs[R_CS].flags >> DESC_DPL_SHIFT & 3) != 0) {
323         env->segs[R_CS].flags &= ~(env->segs[R_CS].flags & DESC_DPL_MASK);
324         env->segs[R_DS].flags &= ~(env->segs[R_DS].flags & DESC_DPL_MASK);
325         env->segs[R_ES].flags &= ~(env->segs[R_ES].flags & DESC_DPL_MASK);
326         env->segs[R_FS].flags &= ~(env->segs[R_FS].flags & DESC_DPL_MASK);
327         env->segs[R_GS].flags &= ~(env->segs[R_GS].flags & DESC_DPL_MASK);
328         env->segs[R_SS].flags &= ~(env->segs[R_SS].flags & DESC_DPL_MASK);
329     }
330 
331     /* Older versions of QEMU incorrectly used CS.DPL as the CPL when
332      * running under KVM.  This is wrong for conforming code segments.
333      * Luckily, in our implementation the CPL field of hflags is redundant
334      * and we can get the right value from the SS descriptor privilege level.
335      */
336     env->hflags &= ~HF_CPL_MASK;
337     env->hflags |= (env->segs[R_SS].flags >> DESC_DPL_SHIFT) & HF_CPL_MASK;
338 
339 #ifdef CONFIG_KVM
340     if ((env->hflags & HF_GUEST_MASK) &&
341         (!env->nested_state ||
342         !(env->nested_state->flags & KVM_STATE_NESTED_GUEST_MODE))) {
343         error_report("vCPU set in guest-mode inconsistent with "
344                      "migrated kernel nested state");
345         return -EINVAL;
346     }
347 #endif
348 
349     /*
350      * There are cases that we can get valid exception_nr with both
351      * exception_pending and exception_injected being cleared.
352      * This can happen in one of the following scenarios:
353      * 1) Source is older QEMU without KVM_CAP_EXCEPTION_PAYLOAD support.
354      * 2) Source is running on kernel without KVM_CAP_EXCEPTION_PAYLOAD support.
355      * 3) "cpu/exception_info" subsection not sent because there is no exception
356      *    pending or guest wasn't running L2 (See comment in cpu_pre_save()).
357      *
358      * In those cases, we can just deduce that a valid exception_nr means
359      * we can treat the exception as already injected.
360      */
361     if ((env->exception_nr != -1) &&
362         !env->exception_pending && !env->exception_injected) {
363         env->exception_injected = 1;
364     }
365 
366     env->fpstt = (env->fpus_vmstate >> 11) & 7;
367     env->fpus = env->fpus_vmstate & ~0x3800;
368     env->fptag_vmstate ^= 0xff;
369     for(i = 0; i < 8; i++) {
370         env->fptags[i] = (env->fptag_vmstate >> i) & 1;
371     }
372     if (tcg_enabled()) {
373         target_ulong dr7;
374         update_fp_status(env);
375         update_mxcsr_status(env);
376 
377         cpu_breakpoint_remove_all(cs, BP_CPU);
378         cpu_watchpoint_remove_all(cs, BP_CPU);
379 
380         /* Indicate all breakpoints disabled, as they are, then
381            let the helper re-enable them.  */
382         dr7 = env->dr[7];
383         env->dr[7] = dr7 & ~(DR7_GLOBAL_BP_MASK | DR7_LOCAL_BP_MASK);
384         cpu_x86_update_dr7(env, dr7);
385     }
386     tlb_flush(cs);
387     return 0;
388 }
389 
390 static bool async_pf_msr_needed(void *opaque)
391 {
392     X86CPU *cpu = opaque;
393 
394     return cpu->env.async_pf_en_msr != 0;
395 }
396 
397 static bool async_pf_int_msr_needed(void *opaque)
398 {
399     X86CPU *cpu = opaque;
400 
401     return cpu->env.async_pf_int_msr != 0;
402 }
403 
404 static bool pv_eoi_msr_needed(void *opaque)
405 {
406     X86CPU *cpu = opaque;
407 
408     return cpu->env.pv_eoi_en_msr != 0;
409 }
410 
411 static bool steal_time_msr_needed(void *opaque)
412 {
413     X86CPU *cpu = opaque;
414 
415     return cpu->env.steal_time_msr != 0;
416 }
417 
418 static bool exception_info_needed(void *opaque)
419 {
420     X86CPU *cpu = opaque;
421     CPUX86State *env = &cpu->env;
422 
423     /*
424      * It is important to save exception-info only in case
425      * we need to distinguish between a pending and injected
426      * exception. Which is only required in case there is a
427      * pending exception and vCPU is running L2.
428      * For more info, refer to comment in cpu_pre_save().
429      */
430     return env->exception_pending && (env->hflags & HF_GUEST_MASK);
431 }
432 
433 static const VMStateDescription vmstate_exception_info = {
434     .name = "cpu/exception_info",
435     .version_id = 1,
436     .minimum_version_id = 1,
437     .needed = exception_info_needed,
438     .fields = (VMStateField[]) {
439         VMSTATE_UINT8(env.exception_pending, X86CPU),
440         VMSTATE_UINT8(env.exception_injected, X86CPU),
441         VMSTATE_UINT8(env.exception_has_payload, X86CPU),
442         VMSTATE_UINT64(env.exception_payload, X86CPU),
443         VMSTATE_END_OF_LIST()
444     }
445 };
446 
447 /* Poll control MSR enabled by default */
448 static bool poll_control_msr_needed(void *opaque)
449 {
450     X86CPU *cpu = opaque;
451 
452     return cpu->env.poll_control_msr != 1;
453 }
454 
455 static const VMStateDescription vmstate_steal_time_msr = {
456     .name = "cpu/steal_time_msr",
457     .version_id = 1,
458     .minimum_version_id = 1,
459     .needed = steal_time_msr_needed,
460     .fields = (VMStateField[]) {
461         VMSTATE_UINT64(env.steal_time_msr, X86CPU),
462         VMSTATE_END_OF_LIST()
463     }
464 };
465 
466 static const VMStateDescription vmstate_async_pf_msr = {
467     .name = "cpu/async_pf_msr",
468     .version_id = 1,
469     .minimum_version_id = 1,
470     .needed = async_pf_msr_needed,
471     .fields = (VMStateField[]) {
472         VMSTATE_UINT64(env.async_pf_en_msr, X86CPU),
473         VMSTATE_END_OF_LIST()
474     }
475 };
476 
477 static const VMStateDescription vmstate_async_pf_int_msr = {
478     .name = "cpu/async_pf_int_msr",
479     .version_id = 1,
480     .minimum_version_id = 1,
481     .needed = async_pf_int_msr_needed,
482     .fields = (VMStateField[]) {
483         VMSTATE_UINT64(env.async_pf_int_msr, X86CPU),
484         VMSTATE_END_OF_LIST()
485     }
486 };
487 
488 static const VMStateDescription vmstate_pv_eoi_msr = {
489     .name = "cpu/async_pv_eoi_msr",
490     .version_id = 1,
491     .minimum_version_id = 1,
492     .needed = pv_eoi_msr_needed,
493     .fields = (VMStateField[]) {
494         VMSTATE_UINT64(env.pv_eoi_en_msr, X86CPU),
495         VMSTATE_END_OF_LIST()
496     }
497 };
498 
499 static const VMStateDescription vmstate_poll_control_msr = {
500     .name = "cpu/poll_control_msr",
501     .version_id = 1,
502     .minimum_version_id = 1,
503     .needed = poll_control_msr_needed,
504     .fields = (VMStateField[]) {
505         VMSTATE_UINT64(env.poll_control_msr, X86CPU),
506         VMSTATE_END_OF_LIST()
507     }
508 };
509 
510 static bool fpop_ip_dp_needed(void *opaque)
511 {
512     X86CPU *cpu = opaque;
513     CPUX86State *env = &cpu->env;
514 
515     return env->fpop != 0 || env->fpip != 0 || env->fpdp != 0;
516 }
517 
518 static const VMStateDescription vmstate_fpop_ip_dp = {
519     .name = "cpu/fpop_ip_dp",
520     .version_id = 1,
521     .minimum_version_id = 1,
522     .needed = fpop_ip_dp_needed,
523     .fields = (VMStateField[]) {
524         VMSTATE_UINT16(env.fpop, X86CPU),
525         VMSTATE_UINT64(env.fpip, X86CPU),
526         VMSTATE_UINT64(env.fpdp, X86CPU),
527         VMSTATE_END_OF_LIST()
528     }
529 };
530 
531 static bool tsc_adjust_needed(void *opaque)
532 {
533     X86CPU *cpu = opaque;
534     CPUX86State *env = &cpu->env;
535 
536     return env->tsc_adjust != 0;
537 }
538 
539 static const VMStateDescription vmstate_msr_tsc_adjust = {
540     .name = "cpu/msr_tsc_adjust",
541     .version_id = 1,
542     .minimum_version_id = 1,
543     .needed = tsc_adjust_needed,
544     .fields = (VMStateField[]) {
545         VMSTATE_UINT64(env.tsc_adjust, X86CPU),
546         VMSTATE_END_OF_LIST()
547     }
548 };
549 
550 static bool msr_smi_count_needed(void *opaque)
551 {
552     X86CPU *cpu = opaque;
553     CPUX86State *env = &cpu->env;
554 
555     return cpu->migrate_smi_count && env->msr_smi_count != 0;
556 }
557 
558 static const VMStateDescription vmstate_msr_smi_count = {
559     .name = "cpu/msr_smi_count",
560     .version_id = 1,
561     .minimum_version_id = 1,
562     .needed = msr_smi_count_needed,
563     .fields = (VMStateField[]) {
564         VMSTATE_UINT64(env.msr_smi_count, X86CPU),
565         VMSTATE_END_OF_LIST()
566     }
567 };
568 
569 static bool tscdeadline_needed(void *opaque)
570 {
571     X86CPU *cpu = opaque;
572     CPUX86State *env = &cpu->env;
573 
574     return env->tsc_deadline != 0;
575 }
576 
577 static const VMStateDescription vmstate_msr_tscdeadline = {
578     .name = "cpu/msr_tscdeadline",
579     .version_id = 1,
580     .minimum_version_id = 1,
581     .needed = tscdeadline_needed,
582     .fields = (VMStateField[]) {
583         VMSTATE_UINT64(env.tsc_deadline, X86CPU),
584         VMSTATE_END_OF_LIST()
585     }
586 };
587 
588 static bool misc_enable_needed(void *opaque)
589 {
590     X86CPU *cpu = opaque;
591     CPUX86State *env = &cpu->env;
592 
593     return env->msr_ia32_misc_enable != MSR_IA32_MISC_ENABLE_DEFAULT;
594 }
595 
596 static bool feature_control_needed(void *opaque)
597 {
598     X86CPU *cpu = opaque;
599     CPUX86State *env = &cpu->env;
600 
601     return env->msr_ia32_feature_control != 0;
602 }
603 
604 static const VMStateDescription vmstate_msr_ia32_misc_enable = {
605     .name = "cpu/msr_ia32_misc_enable",
606     .version_id = 1,
607     .minimum_version_id = 1,
608     .needed = misc_enable_needed,
609     .fields = (VMStateField[]) {
610         VMSTATE_UINT64(env.msr_ia32_misc_enable, X86CPU),
611         VMSTATE_END_OF_LIST()
612     }
613 };
614 
615 static const VMStateDescription vmstate_msr_ia32_feature_control = {
616     .name = "cpu/msr_ia32_feature_control",
617     .version_id = 1,
618     .minimum_version_id = 1,
619     .needed = feature_control_needed,
620     .fields = (VMStateField[]) {
621         VMSTATE_UINT64(env.msr_ia32_feature_control, X86CPU),
622         VMSTATE_END_OF_LIST()
623     }
624 };
625 
626 static bool pmu_enable_needed(void *opaque)
627 {
628     X86CPU *cpu = opaque;
629     CPUX86State *env = &cpu->env;
630     int i;
631 
632     if (env->msr_fixed_ctr_ctrl || env->msr_global_ctrl ||
633         env->msr_global_status || env->msr_global_ovf_ctrl) {
634         return true;
635     }
636     for (i = 0; i < MAX_FIXED_COUNTERS; i++) {
637         if (env->msr_fixed_counters[i]) {
638             return true;
639         }
640     }
641     for (i = 0; i < MAX_GP_COUNTERS; i++) {
642         if (env->msr_gp_counters[i] || env->msr_gp_evtsel[i]) {
643             return true;
644         }
645     }
646 
647     return false;
648 }
649 
650 static const VMStateDescription vmstate_msr_architectural_pmu = {
651     .name = "cpu/msr_architectural_pmu",
652     .version_id = 1,
653     .minimum_version_id = 1,
654     .needed = pmu_enable_needed,
655     .fields = (VMStateField[]) {
656         VMSTATE_UINT64(env.msr_fixed_ctr_ctrl, X86CPU),
657         VMSTATE_UINT64(env.msr_global_ctrl, X86CPU),
658         VMSTATE_UINT64(env.msr_global_status, X86CPU),
659         VMSTATE_UINT64(env.msr_global_ovf_ctrl, X86CPU),
660         VMSTATE_UINT64_ARRAY(env.msr_fixed_counters, X86CPU, MAX_FIXED_COUNTERS),
661         VMSTATE_UINT64_ARRAY(env.msr_gp_counters, X86CPU, MAX_GP_COUNTERS),
662         VMSTATE_UINT64_ARRAY(env.msr_gp_evtsel, X86CPU, MAX_GP_COUNTERS),
663         VMSTATE_END_OF_LIST()
664     }
665 };
666 
667 static bool mpx_needed(void *opaque)
668 {
669     X86CPU *cpu = opaque;
670     CPUX86State *env = &cpu->env;
671     unsigned int i;
672 
673     for (i = 0; i < 4; i++) {
674         if (env->bnd_regs[i].lb || env->bnd_regs[i].ub) {
675             return true;
676         }
677     }
678 
679     if (env->bndcs_regs.cfgu || env->bndcs_regs.sts) {
680         return true;
681     }
682 
683     return !!env->msr_bndcfgs;
684 }
685 
686 static const VMStateDescription vmstate_mpx = {
687     .name = "cpu/mpx",
688     .version_id = 1,
689     .minimum_version_id = 1,
690     .needed = mpx_needed,
691     .fields = (VMStateField[]) {
692         VMSTATE_BND_REGS(env.bnd_regs, X86CPU, 4),
693         VMSTATE_UINT64(env.bndcs_regs.cfgu, X86CPU),
694         VMSTATE_UINT64(env.bndcs_regs.sts, X86CPU),
695         VMSTATE_UINT64(env.msr_bndcfgs, X86CPU),
696         VMSTATE_END_OF_LIST()
697     }
698 };
699 
700 static bool hyperv_hypercall_enable_needed(void *opaque)
701 {
702     X86CPU *cpu = opaque;
703     CPUX86State *env = &cpu->env;
704 
705     return env->msr_hv_hypercall != 0 || env->msr_hv_guest_os_id != 0;
706 }
707 
708 static const VMStateDescription vmstate_msr_hyperv_hypercall = {
709     .name = "cpu/msr_hyperv_hypercall",
710     .version_id = 1,
711     .minimum_version_id = 1,
712     .needed = hyperv_hypercall_enable_needed,
713     .fields = (VMStateField[]) {
714         VMSTATE_UINT64(env.msr_hv_guest_os_id, X86CPU),
715         VMSTATE_UINT64(env.msr_hv_hypercall, X86CPU),
716         VMSTATE_END_OF_LIST()
717     }
718 };
719 
720 static bool hyperv_vapic_enable_needed(void *opaque)
721 {
722     X86CPU *cpu = opaque;
723     CPUX86State *env = &cpu->env;
724 
725     return env->msr_hv_vapic != 0;
726 }
727 
728 static const VMStateDescription vmstate_msr_hyperv_vapic = {
729     .name = "cpu/msr_hyperv_vapic",
730     .version_id = 1,
731     .minimum_version_id = 1,
732     .needed = hyperv_vapic_enable_needed,
733     .fields = (VMStateField[]) {
734         VMSTATE_UINT64(env.msr_hv_vapic, X86CPU),
735         VMSTATE_END_OF_LIST()
736     }
737 };
738 
739 static bool hyperv_time_enable_needed(void *opaque)
740 {
741     X86CPU *cpu = opaque;
742     CPUX86State *env = &cpu->env;
743 
744     return env->msr_hv_tsc != 0;
745 }
746 
747 static const VMStateDescription vmstate_msr_hyperv_time = {
748     .name = "cpu/msr_hyperv_time",
749     .version_id = 1,
750     .minimum_version_id = 1,
751     .needed = hyperv_time_enable_needed,
752     .fields = (VMStateField[]) {
753         VMSTATE_UINT64(env.msr_hv_tsc, X86CPU),
754         VMSTATE_END_OF_LIST()
755     }
756 };
757 
758 static bool hyperv_crash_enable_needed(void *opaque)
759 {
760     X86CPU *cpu = opaque;
761     CPUX86State *env = &cpu->env;
762     int i;
763 
764     for (i = 0; i < HV_CRASH_PARAMS; i++) {
765         if (env->msr_hv_crash_params[i]) {
766             return true;
767         }
768     }
769     return false;
770 }
771 
772 static const VMStateDescription vmstate_msr_hyperv_crash = {
773     .name = "cpu/msr_hyperv_crash",
774     .version_id = 1,
775     .minimum_version_id = 1,
776     .needed = hyperv_crash_enable_needed,
777     .fields = (VMStateField[]) {
778         VMSTATE_UINT64_ARRAY(env.msr_hv_crash_params, X86CPU, HV_CRASH_PARAMS),
779         VMSTATE_END_OF_LIST()
780     }
781 };
782 
783 static bool hyperv_runtime_enable_needed(void *opaque)
784 {
785     X86CPU *cpu = opaque;
786     CPUX86State *env = &cpu->env;
787 
788     if (!hyperv_feat_enabled(cpu, HYPERV_FEAT_RUNTIME)) {
789         return false;
790     }
791 
792     return env->msr_hv_runtime != 0;
793 }
794 
795 static const VMStateDescription vmstate_msr_hyperv_runtime = {
796     .name = "cpu/msr_hyperv_runtime",
797     .version_id = 1,
798     .minimum_version_id = 1,
799     .needed = hyperv_runtime_enable_needed,
800     .fields = (VMStateField[]) {
801         VMSTATE_UINT64(env.msr_hv_runtime, X86CPU),
802         VMSTATE_END_OF_LIST()
803     }
804 };
805 
806 static bool hyperv_synic_enable_needed(void *opaque)
807 {
808     X86CPU *cpu = opaque;
809     CPUX86State *env = &cpu->env;
810     int i;
811 
812     if (env->msr_hv_synic_control != 0 ||
813         env->msr_hv_synic_evt_page != 0 ||
814         env->msr_hv_synic_msg_page != 0) {
815         return true;
816     }
817 
818     for (i = 0; i < ARRAY_SIZE(env->msr_hv_synic_sint); i++) {
819         if (env->msr_hv_synic_sint[i] != 0) {
820             return true;
821         }
822     }
823 
824     return false;
825 }
826 
827 static int hyperv_synic_post_load(void *opaque, int version_id)
828 {
829     X86CPU *cpu = opaque;
830     hyperv_x86_synic_update(cpu);
831     return 0;
832 }
833 
834 static const VMStateDescription vmstate_msr_hyperv_synic = {
835     .name = "cpu/msr_hyperv_synic",
836     .version_id = 1,
837     .minimum_version_id = 1,
838     .needed = hyperv_synic_enable_needed,
839     .post_load = hyperv_synic_post_load,
840     .fields = (VMStateField[]) {
841         VMSTATE_UINT64(env.msr_hv_synic_control, X86CPU),
842         VMSTATE_UINT64(env.msr_hv_synic_evt_page, X86CPU),
843         VMSTATE_UINT64(env.msr_hv_synic_msg_page, X86CPU),
844         VMSTATE_UINT64_ARRAY(env.msr_hv_synic_sint, X86CPU, HV_SINT_COUNT),
845         VMSTATE_END_OF_LIST()
846     }
847 };
848 
849 static bool hyperv_stimer_enable_needed(void *opaque)
850 {
851     X86CPU *cpu = opaque;
852     CPUX86State *env = &cpu->env;
853     int i;
854 
855     for (i = 0; i < ARRAY_SIZE(env->msr_hv_stimer_config); i++) {
856         if (env->msr_hv_stimer_config[i] || env->msr_hv_stimer_count[i]) {
857             return true;
858         }
859     }
860     return false;
861 }
862 
863 static const VMStateDescription vmstate_msr_hyperv_stimer = {
864     .name = "cpu/msr_hyperv_stimer",
865     .version_id = 1,
866     .minimum_version_id = 1,
867     .needed = hyperv_stimer_enable_needed,
868     .fields = (VMStateField[]) {
869         VMSTATE_UINT64_ARRAY(env.msr_hv_stimer_config, X86CPU,
870                              HV_STIMER_COUNT),
871         VMSTATE_UINT64_ARRAY(env.msr_hv_stimer_count, X86CPU, HV_STIMER_COUNT),
872         VMSTATE_END_OF_LIST()
873     }
874 };
875 
876 static bool hyperv_reenlightenment_enable_needed(void *opaque)
877 {
878     X86CPU *cpu = opaque;
879     CPUX86State *env = &cpu->env;
880 
881     return env->msr_hv_reenlightenment_control != 0 ||
882         env->msr_hv_tsc_emulation_control != 0 ||
883         env->msr_hv_tsc_emulation_status != 0;
884 }
885 
886 static int hyperv_reenlightenment_post_load(void *opaque, int version_id)
887 {
888     X86CPU *cpu = opaque;
889     CPUX86State *env = &cpu->env;
890 
891     /*
892      * KVM doesn't fully support re-enlightenment notifications so we need to
893      * make sure TSC frequency doesn't change upon migration.
894      */
895     if ((env->msr_hv_reenlightenment_control & HV_REENLIGHTENMENT_ENABLE_BIT) &&
896         !env->user_tsc_khz) {
897         error_report("Guest enabled re-enlightenment notifications, "
898                      "'tsc-frequency=' has to be specified");
899         return -EINVAL;
900     }
901 
902     return 0;
903 }
904 
905 static const VMStateDescription vmstate_msr_hyperv_reenlightenment = {
906     .name = "cpu/msr_hyperv_reenlightenment",
907     .version_id = 1,
908     .minimum_version_id = 1,
909     .needed = hyperv_reenlightenment_enable_needed,
910     .post_load = hyperv_reenlightenment_post_load,
911     .fields = (VMStateField[]) {
912         VMSTATE_UINT64(env.msr_hv_reenlightenment_control, X86CPU),
913         VMSTATE_UINT64(env.msr_hv_tsc_emulation_control, X86CPU),
914         VMSTATE_UINT64(env.msr_hv_tsc_emulation_status, X86CPU),
915         VMSTATE_END_OF_LIST()
916     }
917 };
918 
919 static bool avx512_needed(void *opaque)
920 {
921     X86CPU *cpu = opaque;
922     CPUX86State *env = &cpu->env;
923     unsigned int i;
924 
925     for (i = 0; i < NB_OPMASK_REGS; i++) {
926         if (env->opmask_regs[i]) {
927             return true;
928         }
929     }
930 
931     for (i = 0; i < CPU_NB_REGS; i++) {
932 #define ENV_XMM(reg, field) (env->xmm_regs[reg].ZMM_Q(field))
933         if (ENV_XMM(i, 4) || ENV_XMM(i, 6) ||
934             ENV_XMM(i, 5) || ENV_XMM(i, 7)) {
935             return true;
936         }
937 #ifdef TARGET_X86_64
938         if (ENV_XMM(i+16, 0) || ENV_XMM(i+16, 1) ||
939             ENV_XMM(i+16, 2) || ENV_XMM(i+16, 3) ||
940             ENV_XMM(i+16, 4) || ENV_XMM(i+16, 5) ||
941             ENV_XMM(i+16, 6) || ENV_XMM(i+16, 7)) {
942             return true;
943         }
944 #endif
945     }
946 
947     return false;
948 }
949 
950 static const VMStateDescription vmstate_avx512 = {
951     .name = "cpu/avx512",
952     .version_id = 1,
953     .minimum_version_id = 1,
954     .needed = avx512_needed,
955     .fields = (VMStateField[]) {
956         VMSTATE_UINT64_ARRAY(env.opmask_regs, X86CPU, NB_OPMASK_REGS),
957         VMSTATE_ZMMH_REGS_VARS(env.xmm_regs, X86CPU, 0),
958 #ifdef TARGET_X86_64
959         VMSTATE_Hi16_ZMM_REGS_VARS(env.xmm_regs, X86CPU, 16),
960 #endif
961         VMSTATE_END_OF_LIST()
962     }
963 };
964 
965 static bool xss_needed(void *opaque)
966 {
967     X86CPU *cpu = opaque;
968     CPUX86State *env = &cpu->env;
969 
970     return env->xss != 0;
971 }
972 
973 static const VMStateDescription vmstate_xss = {
974     .name = "cpu/xss",
975     .version_id = 1,
976     .minimum_version_id = 1,
977     .needed = xss_needed,
978     .fields = (VMStateField[]) {
979         VMSTATE_UINT64(env.xss, X86CPU),
980         VMSTATE_END_OF_LIST()
981     }
982 };
983 
984 static bool umwait_needed(void *opaque)
985 {
986     X86CPU *cpu = opaque;
987     CPUX86State *env = &cpu->env;
988 
989     return env->umwait != 0;
990 }
991 
992 static const VMStateDescription vmstate_umwait = {
993     .name = "cpu/umwait",
994     .version_id = 1,
995     .minimum_version_id = 1,
996     .needed = umwait_needed,
997     .fields = (VMStateField[]) {
998         VMSTATE_UINT32(env.umwait, X86CPU),
999         VMSTATE_END_OF_LIST()
1000     }
1001 };
1002 
1003 static bool pkru_needed(void *opaque)
1004 {
1005     X86CPU *cpu = opaque;
1006     CPUX86State *env = &cpu->env;
1007 
1008     return env->pkru != 0;
1009 }
1010 
1011 static const VMStateDescription vmstate_pkru = {
1012     .name = "cpu/pkru",
1013     .version_id = 1,
1014     .minimum_version_id = 1,
1015     .needed = pkru_needed,
1016     .fields = (VMStateField[]){
1017         VMSTATE_UINT32(env.pkru, X86CPU),
1018         VMSTATE_END_OF_LIST()
1019     }
1020 };
1021 
1022 static bool pkrs_needed(void *opaque)
1023 {
1024     X86CPU *cpu = opaque;
1025     CPUX86State *env = &cpu->env;
1026 
1027     return env->pkrs != 0;
1028 }
1029 
1030 static const VMStateDescription vmstate_pkrs = {
1031     .name = "cpu/pkrs",
1032     .version_id = 1,
1033     .minimum_version_id = 1,
1034     .needed = pkrs_needed,
1035     .fields = (VMStateField[]){
1036         VMSTATE_UINT32(env.pkrs, X86CPU),
1037         VMSTATE_END_OF_LIST()
1038     }
1039 };
1040 
1041 static bool tsc_khz_needed(void *opaque)
1042 {
1043     X86CPU *cpu = opaque;
1044     CPUX86State *env = &cpu->env;
1045     MachineClass *mc = MACHINE_GET_CLASS(qdev_get_machine());
1046     X86MachineClass *x86mc = X86_MACHINE_CLASS(mc);
1047     return env->tsc_khz && x86mc->save_tsc_khz;
1048 }
1049 
1050 static const VMStateDescription vmstate_tsc_khz = {
1051     .name = "cpu/tsc_khz",
1052     .version_id = 1,
1053     .minimum_version_id = 1,
1054     .needed = tsc_khz_needed,
1055     .fields = (VMStateField[]) {
1056         VMSTATE_INT64(env.tsc_khz, X86CPU),
1057         VMSTATE_END_OF_LIST()
1058     }
1059 };
1060 
1061 #ifdef CONFIG_KVM
1062 
1063 static bool vmx_vmcs12_needed(void *opaque)
1064 {
1065     struct kvm_nested_state *nested_state = opaque;
1066     return (nested_state->size >
1067             offsetof(struct kvm_nested_state, data.vmx[0].vmcs12));
1068 }
1069 
1070 static const VMStateDescription vmstate_vmx_vmcs12 = {
1071     .name = "cpu/kvm_nested_state/vmx/vmcs12",
1072     .version_id = 1,
1073     .minimum_version_id = 1,
1074     .needed = vmx_vmcs12_needed,
1075     .fields = (VMStateField[]) {
1076         VMSTATE_UINT8_ARRAY(data.vmx[0].vmcs12,
1077                             struct kvm_nested_state,
1078                             KVM_STATE_NESTED_VMX_VMCS_SIZE),
1079         VMSTATE_END_OF_LIST()
1080     }
1081 };
1082 
1083 static bool vmx_shadow_vmcs12_needed(void *opaque)
1084 {
1085     struct kvm_nested_state *nested_state = opaque;
1086     return (nested_state->size >
1087             offsetof(struct kvm_nested_state, data.vmx[0].shadow_vmcs12));
1088 }
1089 
1090 static const VMStateDescription vmstate_vmx_shadow_vmcs12 = {
1091     .name = "cpu/kvm_nested_state/vmx/shadow_vmcs12",
1092     .version_id = 1,
1093     .minimum_version_id = 1,
1094     .needed = vmx_shadow_vmcs12_needed,
1095     .fields = (VMStateField[]) {
1096         VMSTATE_UINT8_ARRAY(data.vmx[0].shadow_vmcs12,
1097                             struct kvm_nested_state,
1098                             KVM_STATE_NESTED_VMX_VMCS_SIZE),
1099         VMSTATE_END_OF_LIST()
1100     }
1101 };
1102 
1103 static bool vmx_nested_state_needed(void *opaque)
1104 {
1105     struct kvm_nested_state *nested_state = opaque;
1106 
1107     return (nested_state->format == KVM_STATE_NESTED_FORMAT_VMX &&
1108             nested_state->hdr.vmx.vmxon_pa != -1ull);
1109 }
1110 
1111 static const VMStateDescription vmstate_vmx_nested_state = {
1112     .name = "cpu/kvm_nested_state/vmx",
1113     .version_id = 1,
1114     .minimum_version_id = 1,
1115     .needed = vmx_nested_state_needed,
1116     .fields = (VMStateField[]) {
1117         VMSTATE_U64(hdr.vmx.vmxon_pa, struct kvm_nested_state),
1118         VMSTATE_U64(hdr.vmx.vmcs12_pa, struct kvm_nested_state),
1119         VMSTATE_U16(hdr.vmx.smm.flags, struct kvm_nested_state),
1120         VMSTATE_END_OF_LIST()
1121     },
1122     .subsections = (const VMStateDescription*[]) {
1123         &vmstate_vmx_vmcs12,
1124         &vmstate_vmx_shadow_vmcs12,
1125         NULL,
1126     }
1127 };
1128 
1129 static bool svm_nested_state_needed(void *opaque)
1130 {
1131     struct kvm_nested_state *nested_state = opaque;
1132 
1133     /*
1134      * HF_GUEST_MASK and HF2_GIF_MASK are already serialized
1135      * via hflags and hflags2, all that's left is the opaque
1136      * nested state blob.
1137      */
1138     return (nested_state->format == KVM_STATE_NESTED_FORMAT_SVM &&
1139             nested_state->size > offsetof(struct kvm_nested_state, data));
1140 }
1141 
1142 static const VMStateDescription vmstate_svm_nested_state = {
1143     .name = "cpu/kvm_nested_state/svm",
1144     .version_id = 1,
1145     .minimum_version_id = 1,
1146     .needed = svm_nested_state_needed,
1147     .fields = (VMStateField[]) {
1148         VMSTATE_U64(hdr.svm.vmcb_pa, struct kvm_nested_state),
1149         VMSTATE_UINT8_ARRAY(data.svm[0].vmcb12,
1150                             struct kvm_nested_state,
1151                             KVM_STATE_NESTED_SVM_VMCB_SIZE),
1152         VMSTATE_END_OF_LIST()
1153     }
1154 };
1155 
1156 static bool nested_state_needed(void *opaque)
1157 {
1158     X86CPU *cpu = opaque;
1159     CPUX86State *env = &cpu->env;
1160 
1161     return (env->nested_state &&
1162             (vmx_nested_state_needed(env->nested_state) ||
1163              svm_nested_state_needed(env->nested_state)));
1164 }
1165 
1166 static int nested_state_post_load(void *opaque, int version_id)
1167 {
1168     X86CPU *cpu = opaque;
1169     CPUX86State *env = &cpu->env;
1170     struct kvm_nested_state *nested_state = env->nested_state;
1171     int min_nested_state_len = offsetof(struct kvm_nested_state, data);
1172     int max_nested_state_len = kvm_max_nested_state_length();
1173 
1174     /*
1175      * If our kernel don't support setting nested state
1176      * and we have received nested state from migration stream,
1177      * we need to fail migration
1178      */
1179     if (max_nested_state_len <= 0) {
1180         error_report("Received nested state when kernel cannot restore it");
1181         return -EINVAL;
1182     }
1183 
1184     /*
1185      * Verify that the size of received nested_state struct
1186      * at least cover required header and is not larger
1187      * than the max size that our kernel support
1188      */
1189     if (nested_state->size < min_nested_state_len) {
1190         error_report("Received nested state size less than min: "
1191                      "len=%d, min=%d",
1192                      nested_state->size, min_nested_state_len);
1193         return -EINVAL;
1194     }
1195     if (nested_state->size > max_nested_state_len) {
1196         error_report("Received unsupported nested state size: "
1197                      "nested_state->size=%d, max=%d",
1198                      nested_state->size, max_nested_state_len);
1199         return -EINVAL;
1200     }
1201 
1202     /* Verify format is valid */
1203     if ((nested_state->format != KVM_STATE_NESTED_FORMAT_VMX) &&
1204         (nested_state->format != KVM_STATE_NESTED_FORMAT_SVM)) {
1205         error_report("Received invalid nested state format: %d",
1206                      nested_state->format);
1207         return -EINVAL;
1208     }
1209 
1210     return 0;
1211 }
1212 
1213 static const VMStateDescription vmstate_kvm_nested_state = {
1214     .name = "cpu/kvm_nested_state",
1215     .version_id = 1,
1216     .minimum_version_id = 1,
1217     .fields = (VMStateField[]) {
1218         VMSTATE_U16(flags, struct kvm_nested_state),
1219         VMSTATE_U16(format, struct kvm_nested_state),
1220         VMSTATE_U32(size, struct kvm_nested_state),
1221         VMSTATE_END_OF_LIST()
1222     },
1223     .subsections = (const VMStateDescription*[]) {
1224         &vmstate_vmx_nested_state,
1225         &vmstate_svm_nested_state,
1226         NULL
1227     }
1228 };
1229 
1230 static const VMStateDescription vmstate_nested_state = {
1231     .name = "cpu/nested_state",
1232     .version_id = 1,
1233     .minimum_version_id = 1,
1234     .needed = nested_state_needed,
1235     .post_load = nested_state_post_load,
1236     .fields = (VMStateField[]) {
1237         VMSTATE_STRUCT_POINTER(env.nested_state, X86CPU,
1238                 vmstate_kvm_nested_state,
1239                 struct kvm_nested_state),
1240         VMSTATE_END_OF_LIST()
1241     }
1242 };
1243 
1244 #endif
1245 
1246 static bool mcg_ext_ctl_needed(void *opaque)
1247 {
1248     X86CPU *cpu = opaque;
1249     CPUX86State *env = &cpu->env;
1250     return cpu->enable_lmce && env->mcg_ext_ctl;
1251 }
1252 
1253 static const VMStateDescription vmstate_mcg_ext_ctl = {
1254     .name = "cpu/mcg_ext_ctl",
1255     .version_id = 1,
1256     .minimum_version_id = 1,
1257     .needed = mcg_ext_ctl_needed,
1258     .fields = (VMStateField[]) {
1259         VMSTATE_UINT64(env.mcg_ext_ctl, X86CPU),
1260         VMSTATE_END_OF_LIST()
1261     }
1262 };
1263 
1264 static bool spec_ctrl_needed(void *opaque)
1265 {
1266     X86CPU *cpu = opaque;
1267     CPUX86State *env = &cpu->env;
1268 
1269     return env->spec_ctrl != 0;
1270 }
1271 
1272 static const VMStateDescription vmstate_spec_ctrl = {
1273     .name = "cpu/spec_ctrl",
1274     .version_id = 1,
1275     .minimum_version_id = 1,
1276     .needed = spec_ctrl_needed,
1277     .fields = (VMStateField[]){
1278         VMSTATE_UINT64(env.spec_ctrl, X86CPU),
1279         VMSTATE_END_OF_LIST()
1280     }
1281 };
1282 
1283 static bool intel_pt_enable_needed(void *opaque)
1284 {
1285     X86CPU *cpu = opaque;
1286     CPUX86State *env = &cpu->env;
1287     int i;
1288 
1289     if (env->msr_rtit_ctrl || env->msr_rtit_status ||
1290         env->msr_rtit_output_base || env->msr_rtit_output_mask ||
1291         env->msr_rtit_cr3_match) {
1292         return true;
1293     }
1294 
1295     for (i = 0; i < MAX_RTIT_ADDRS; i++) {
1296         if (env->msr_rtit_addrs[i]) {
1297             return true;
1298         }
1299     }
1300 
1301     return false;
1302 }
1303 
1304 static const VMStateDescription vmstate_msr_intel_pt = {
1305     .name = "cpu/intel_pt",
1306     .version_id = 1,
1307     .minimum_version_id = 1,
1308     .needed = intel_pt_enable_needed,
1309     .fields = (VMStateField[]) {
1310         VMSTATE_UINT64(env.msr_rtit_ctrl, X86CPU),
1311         VMSTATE_UINT64(env.msr_rtit_status, X86CPU),
1312         VMSTATE_UINT64(env.msr_rtit_output_base, X86CPU),
1313         VMSTATE_UINT64(env.msr_rtit_output_mask, X86CPU),
1314         VMSTATE_UINT64(env.msr_rtit_cr3_match, X86CPU),
1315         VMSTATE_UINT64_ARRAY(env.msr_rtit_addrs, X86CPU, MAX_RTIT_ADDRS),
1316         VMSTATE_END_OF_LIST()
1317     }
1318 };
1319 
1320 static bool virt_ssbd_needed(void *opaque)
1321 {
1322     X86CPU *cpu = opaque;
1323     CPUX86State *env = &cpu->env;
1324 
1325     return env->virt_ssbd != 0;
1326 }
1327 
1328 static const VMStateDescription vmstate_msr_virt_ssbd = {
1329     .name = "cpu/virt_ssbd",
1330     .version_id = 1,
1331     .minimum_version_id = 1,
1332     .needed = virt_ssbd_needed,
1333     .fields = (VMStateField[]){
1334         VMSTATE_UINT64(env.virt_ssbd, X86CPU),
1335         VMSTATE_END_OF_LIST()
1336     }
1337 };
1338 
1339 static bool svm_npt_needed(void *opaque)
1340 {
1341     X86CPU *cpu = opaque;
1342     CPUX86State *env = &cpu->env;
1343 
1344     return !!(env->hflags2 & HF2_NPT_MASK);
1345 }
1346 
1347 static const VMStateDescription vmstate_svm_npt = {
1348     .name = "cpu/svn_npt",
1349     .version_id = 1,
1350     .minimum_version_id = 1,
1351     .needed = svm_npt_needed,
1352     .fields = (VMStateField[]){
1353         VMSTATE_UINT64(env.nested_cr3, X86CPU),
1354         VMSTATE_UINT32(env.nested_pg_mode, X86CPU),
1355         VMSTATE_END_OF_LIST()
1356     }
1357 };
1358 
1359 static bool svm_guest_needed(void *opaque)
1360 {
1361     X86CPU *cpu = opaque;
1362     CPUX86State *env = &cpu->env;
1363 
1364     return tcg_enabled() && env->int_ctl;
1365 }
1366 
1367 static const VMStateDescription vmstate_svm_guest = {
1368     .name = "cpu/svm_guest",
1369     .version_id = 1,
1370     .minimum_version_id = 1,
1371     .needed = svm_guest_needed,
1372     .fields = (VMStateField[]){
1373         VMSTATE_UINT32(env.int_ctl, X86CPU),
1374         VMSTATE_END_OF_LIST()
1375     }
1376 };
1377 
1378 #ifndef TARGET_X86_64
1379 static bool intel_efer32_needed(void *opaque)
1380 {
1381     X86CPU *cpu = opaque;
1382     CPUX86State *env = &cpu->env;
1383 
1384     return env->efer != 0;
1385 }
1386 
1387 static const VMStateDescription vmstate_efer32 = {
1388     .name = "cpu/efer32",
1389     .version_id = 1,
1390     .minimum_version_id = 1,
1391     .needed = intel_efer32_needed,
1392     .fields = (VMStateField[]) {
1393         VMSTATE_UINT64(env.efer, X86CPU),
1394         VMSTATE_END_OF_LIST()
1395     }
1396 };
1397 #endif
1398 
1399 static bool msr_tsx_ctrl_needed(void *opaque)
1400 {
1401     X86CPU *cpu = opaque;
1402     CPUX86State *env = &cpu->env;
1403 
1404     return env->features[FEAT_ARCH_CAPABILITIES] & ARCH_CAP_TSX_CTRL_MSR;
1405 }
1406 
1407 static const VMStateDescription vmstate_msr_tsx_ctrl = {
1408     .name = "cpu/msr_tsx_ctrl",
1409     .version_id = 1,
1410     .minimum_version_id = 1,
1411     .needed = msr_tsx_ctrl_needed,
1412     .fields = (VMStateField[]) {
1413         VMSTATE_UINT32(env.tsx_ctrl, X86CPU),
1414         VMSTATE_END_OF_LIST()
1415     }
1416 };
1417 
1418 const VMStateDescription vmstate_x86_cpu = {
1419     .name = "cpu",
1420     .version_id = 12,
1421     .minimum_version_id = 11,
1422     .pre_save = cpu_pre_save,
1423     .post_load = cpu_post_load,
1424     .fields = (VMStateField[]) {
1425         VMSTATE_UINTTL_ARRAY(env.regs, X86CPU, CPU_NB_REGS),
1426         VMSTATE_UINTTL(env.eip, X86CPU),
1427         VMSTATE_UINTTL(env.eflags, X86CPU),
1428         VMSTATE_UINT32(env.hflags, X86CPU),
1429         /* FPU */
1430         VMSTATE_UINT16(env.fpuc, X86CPU),
1431         VMSTATE_UINT16(env.fpus_vmstate, X86CPU),
1432         VMSTATE_UINT16(env.fptag_vmstate, X86CPU),
1433         VMSTATE_UINT16(env.fpregs_format_vmstate, X86CPU),
1434 
1435         VMSTATE_STRUCT_ARRAY(env.fpregs, X86CPU, 8, 0, vmstate_fpreg, FPReg),
1436 
1437         VMSTATE_SEGMENT_ARRAY(env.segs, X86CPU, 6),
1438         VMSTATE_SEGMENT(env.ldt, X86CPU),
1439         VMSTATE_SEGMENT(env.tr, X86CPU),
1440         VMSTATE_SEGMENT(env.gdt, X86CPU),
1441         VMSTATE_SEGMENT(env.idt, X86CPU),
1442 
1443         VMSTATE_UINT32(env.sysenter_cs, X86CPU),
1444         VMSTATE_UINTTL(env.sysenter_esp, X86CPU),
1445         VMSTATE_UINTTL(env.sysenter_eip, X86CPU),
1446 
1447         VMSTATE_UINTTL(env.cr[0], X86CPU),
1448         VMSTATE_UINTTL(env.cr[2], X86CPU),
1449         VMSTATE_UINTTL(env.cr[3], X86CPU),
1450         VMSTATE_UINTTL(env.cr[4], X86CPU),
1451         VMSTATE_UINTTL_ARRAY(env.dr, X86CPU, 8),
1452         /* MMU */
1453         VMSTATE_INT32(env.a20_mask, X86CPU),
1454         /* XMM */
1455         VMSTATE_UINT32(env.mxcsr, X86CPU),
1456         VMSTATE_XMM_REGS(env.xmm_regs, X86CPU, 0),
1457 
1458 #ifdef TARGET_X86_64
1459         VMSTATE_UINT64(env.efer, X86CPU),
1460         VMSTATE_UINT64(env.star, X86CPU),
1461         VMSTATE_UINT64(env.lstar, X86CPU),
1462         VMSTATE_UINT64(env.cstar, X86CPU),
1463         VMSTATE_UINT64(env.fmask, X86CPU),
1464         VMSTATE_UINT64(env.kernelgsbase, X86CPU),
1465 #endif
1466         VMSTATE_UINT32(env.smbase, X86CPU),
1467 
1468         VMSTATE_UINT64(env.pat, X86CPU),
1469         VMSTATE_UINT32(env.hflags2, X86CPU),
1470 
1471         VMSTATE_UINT64(env.vm_hsave, X86CPU),
1472         VMSTATE_UINT64(env.vm_vmcb, X86CPU),
1473         VMSTATE_UINT64(env.tsc_offset, X86CPU),
1474         VMSTATE_UINT64(env.intercept, X86CPU),
1475         VMSTATE_UINT16(env.intercept_cr_read, X86CPU),
1476         VMSTATE_UINT16(env.intercept_cr_write, X86CPU),
1477         VMSTATE_UINT16(env.intercept_dr_read, X86CPU),
1478         VMSTATE_UINT16(env.intercept_dr_write, X86CPU),
1479         VMSTATE_UINT32(env.intercept_exceptions, X86CPU),
1480         VMSTATE_UINT8(env.v_tpr, X86CPU),
1481         /* MTRRs */
1482         VMSTATE_UINT64_ARRAY(env.mtrr_fixed, X86CPU, 11),
1483         VMSTATE_UINT64(env.mtrr_deftype, X86CPU),
1484         VMSTATE_MTRR_VARS(env.mtrr_var, X86CPU, MSR_MTRRcap_VCNT, 8),
1485         /* KVM-related states */
1486         VMSTATE_INT32(env.interrupt_injected, X86CPU),
1487         VMSTATE_UINT32(env.mp_state, X86CPU),
1488         VMSTATE_UINT64(env.tsc, X86CPU),
1489         VMSTATE_INT32(env.exception_nr, X86CPU),
1490         VMSTATE_UINT8(env.soft_interrupt, X86CPU),
1491         VMSTATE_UINT8(env.nmi_injected, X86CPU),
1492         VMSTATE_UINT8(env.nmi_pending, X86CPU),
1493         VMSTATE_UINT8(env.has_error_code, X86CPU),
1494         VMSTATE_UINT32(env.sipi_vector, X86CPU),
1495         /* MCE */
1496         VMSTATE_UINT64(env.mcg_cap, X86CPU),
1497         VMSTATE_UINT64(env.mcg_status, X86CPU),
1498         VMSTATE_UINT64(env.mcg_ctl, X86CPU),
1499         VMSTATE_UINT64_ARRAY(env.mce_banks, X86CPU, MCE_BANKS_DEF * 4),
1500         /* rdtscp */
1501         VMSTATE_UINT64(env.tsc_aux, X86CPU),
1502         /* KVM pvclock msr */
1503         VMSTATE_UINT64(env.system_time_msr, X86CPU),
1504         VMSTATE_UINT64(env.wall_clock_msr, X86CPU),
1505         /* XSAVE related fields */
1506         VMSTATE_UINT64_V(env.xcr0, X86CPU, 12),
1507         VMSTATE_UINT64_V(env.xstate_bv, X86CPU, 12),
1508         VMSTATE_YMMH_REGS_VARS(env.xmm_regs, X86CPU, 0, 12),
1509         VMSTATE_END_OF_LIST()
1510         /* The above list is not sorted /wrt version numbers, watch out! */
1511     },
1512     .subsections = (const VMStateDescription*[]) {
1513         &vmstate_exception_info,
1514         &vmstate_async_pf_msr,
1515         &vmstate_async_pf_int_msr,
1516         &vmstate_pv_eoi_msr,
1517         &vmstate_steal_time_msr,
1518         &vmstate_poll_control_msr,
1519         &vmstate_fpop_ip_dp,
1520         &vmstate_msr_tsc_adjust,
1521         &vmstate_msr_tscdeadline,
1522         &vmstate_msr_ia32_misc_enable,
1523         &vmstate_msr_ia32_feature_control,
1524         &vmstate_msr_architectural_pmu,
1525         &vmstate_mpx,
1526         &vmstate_msr_hyperv_hypercall,
1527         &vmstate_msr_hyperv_vapic,
1528         &vmstate_msr_hyperv_time,
1529         &vmstate_msr_hyperv_crash,
1530         &vmstate_msr_hyperv_runtime,
1531         &vmstate_msr_hyperv_synic,
1532         &vmstate_msr_hyperv_stimer,
1533         &vmstate_msr_hyperv_reenlightenment,
1534         &vmstate_avx512,
1535         &vmstate_xss,
1536         &vmstate_umwait,
1537         &vmstate_tsc_khz,
1538         &vmstate_msr_smi_count,
1539         &vmstate_pkru,
1540         &vmstate_pkrs,
1541         &vmstate_spec_ctrl,
1542         &vmstate_mcg_ext_ctl,
1543         &vmstate_msr_intel_pt,
1544         &vmstate_msr_virt_ssbd,
1545         &vmstate_svm_npt,
1546         &vmstate_svm_guest,
1547 #ifndef TARGET_X86_64
1548         &vmstate_efer32,
1549 #endif
1550 #ifdef CONFIG_KVM
1551         &vmstate_nested_state,
1552 #endif
1553         &vmstate_msr_tsx_ctrl,
1554         NULL
1555     }
1556 };
1557