summary refs log tree commit diff stats
path: root/target-i386
diff options
context:
space:
mode:
Diffstat (limited to 'target-i386')
-rw-r--r--target-i386/cpu.c11
-rw-r--r--target-i386/cpu.h85
-rw-r--r--target-i386/helper.c14
-rw-r--r--target-i386/kvm.c42
-rw-r--r--target-i386/kvm_i386.h2
-rw-r--r--target-i386/seg_helper.c53
-rw-r--r--target-i386/smm_helper.c34
-rw-r--r--target-i386/svm_helper.c11
-rw-r--r--target-i386/translate.c40
9 files changed, 165 insertions, 127 deletions
diff --git a/target-i386/cpu.c b/target-i386/cpu.c
index 8f193a9330..042a48d703 100644
--- a/target-i386/cpu.c
+++ b/target-i386/cpu.c
@@ -24,6 +24,7 @@
 #include "cpu.h"
 #include "sysemu/kvm.h"
 #include "sysemu/cpus.h"
+#include "kvm_i386.h"
 #include "topology.h"
 
 #include "qemu/option.h"
@@ -2417,8 +2418,7 @@ static void x86_cpu_reset(CPUState *s)
 
     xcc->parent_reset(s);
 
-
-    memset(env, 0, offsetof(CPUX86State, pat));
+    memset(env, 0, offsetof(CPUX86State, cpuid_level));
 
     tlb_flush(s, 1);
 
@@ -2484,8 +2484,7 @@ static void x86_cpu_reset(CPUState *s)
     cpu_breakpoint_remove_all(s, BP_CPU);
     cpu_watchpoint_remove_all(s, BP_CPU);
 
-    env->tsc_adjust = 0;
-    env->tsc = 0;
+    env->xcr0 = 1;
 
 #if !defined(CONFIG_USER_ONLY)
     /* We hard-wire the BSP to the first CPU. */
@@ -2494,6 +2493,10 @@ static void x86_cpu_reset(CPUState *s)
     }
 
     s->halted = !cpu_is_bsp(cpu);
+
+    if (kvm_enabled()) {
+        kvm_arch_reset_vcpu(cpu);
+    }
 #endif
 }
 
diff --git a/target-i386/cpu.h b/target-i386/cpu.h
index 2a22a7d64e..e9cbdabc03 100644
--- a/target-i386/cpu.h
+++ b/target-i386/cpu.h
@@ -124,9 +124,9 @@
 #define ID_MASK                 0x00200000
 
 /* hidden flags - used internally by qemu to represent additional cpu
-   states. Only the CPL, INHIBIT_IRQ, SMM and SVMI are not
-   redundant. We avoid using the IOPL_MASK, TF_MASK, VM_MASK and AC_MASK
-   bit positions to ease oring with eflags. */
+   states. Only the INHIBIT_IRQ, SMM and SVMI are not redundant. We
+   avoid using the IOPL_MASK, TF_MASK, VM_MASK and AC_MASK bit
+   positions to ease oring with eflags. */
 /* current cpl */
 #define HF_CPL_SHIFT         0
 /* true if soft mmu is being used */
@@ -606,10 +606,11 @@ typedef uint32_t FeatureWordArray[FEATURE_WORDS];
 #define CPU_INTERRUPT_NMI       CPU_INTERRUPT_TGT_EXT_3
 #define CPU_INTERRUPT_MCE       CPU_INTERRUPT_TGT_EXT_4
 #define CPU_INTERRUPT_VIRQ      CPU_INTERRUPT_TGT_INT_0
-#define CPU_INTERRUPT_INIT      CPU_INTERRUPT_TGT_INT_1
-#define CPU_INTERRUPT_SIPI      CPU_INTERRUPT_TGT_INT_2
-#define CPU_INTERRUPT_TPR       CPU_INTERRUPT_TGT_INT_3
+#define CPU_INTERRUPT_SIPI      CPU_INTERRUPT_TGT_INT_1
+#define CPU_INTERRUPT_TPR       CPU_INTERRUPT_TGT_INT_2
 
+/* Use a clearer name for this.  */
+#define CPU_INTERRUPT_INIT      CPU_INTERRUPT_RESET
 
 typedef enum {
     CC_OP_DYNAMIC, /* must use dynamic code to get cc_op */
@@ -797,6 +798,13 @@ typedef struct CPUX86State {
     target_ulong cr[5]; /* NOTE: cr1 is unused */
     int32_t a20_mask;
 
+    BNDReg bnd_regs[4];
+    BNDCSReg bndcs_regs;
+    uint64_t msr_bndcfgs;
+
+    /* Beginning of state preserved by INIT (dummy marker).  */
+    struct {} start_init_save;
+
     /* FPU state */
     unsigned int fpstt; /* top of stack index */
     uint16_t fpus;
@@ -819,6 +827,8 @@ typedef struct CPUX86State {
     XMMReg xmm_t0;
     MMXReg mmx_t0;
 
+    XMMReg ymmh_regs[CPU_NB_REGS];
+
     /* sysenter registers */
     uint32_t sysenter_cs;
     target_ulong sysenter_esp;
@@ -827,15 +837,6 @@ typedef struct CPUX86State {
     uint64_t star;
 
     uint64_t vm_hsave;
-    uint64_t vm_vmcb;
-    uint64_t tsc_offset;
-    uint64_t intercept;
-    uint16_t intercept_cr_read;
-    uint16_t intercept_cr_write;
-    uint16_t intercept_dr_read;
-    uint16_t intercept_dr_write;
-    uint32_t intercept_exceptions;
-    uint8_t v_tpr;
 
 #ifdef TARGET_X86_64
     target_ulong lstar;
@@ -843,11 +844,6 @@ typedef struct CPUX86State {
     target_ulong fmask;
     target_ulong kernelgsbase;
 #endif
-    uint64_t system_time_msr;
-    uint64_t wall_clock_msr;
-    uint64_t steal_time_msr;
-    uint64_t async_pf_en_msr;
-    uint64_t pv_eoi_en_msr;
 
     uint64_t tsc;
     uint64_t tsc_adjust;
@@ -864,6 +860,19 @@ typedef struct CPUX86State {
     uint64_t msr_fixed_counters[MAX_FIXED_COUNTERS];
     uint64_t msr_gp_counters[MAX_GP_COUNTERS];
     uint64_t msr_gp_evtsel[MAX_GP_COUNTERS];
+
+    uint64_t pat;
+    uint32_t smbase;
+
+    /* End of state preserved by INIT (dummy marker).  */
+    struct {} end_init_save;
+
+    uint64_t system_time_msr;
+    uint64_t wall_clock_msr;
+    uint64_t steal_time_msr;
+    uint64_t async_pf_en_msr;
+    uint64_t pv_eoi_en_msr;
+
     uint64_t msr_hv_hypercall;
     uint64_t msr_hv_guest_os_id;
     uint64_t msr_hv_vapic;
@@ -878,9 +887,18 @@ typedef struct CPUX86State {
         struct CPUBreakpoint *cpu_breakpoint[4];
         struct CPUWatchpoint *cpu_watchpoint[4];
     }; /* break/watchpoints for dr[0..3] */
-    uint32_t smbase;
     int old_exception;  /* exception in flight */
 
+    uint64_t vm_vmcb;
+    uint64_t tsc_offset;
+    uint64_t intercept;
+    uint16_t intercept_cr_read;
+    uint16_t intercept_cr_write;
+    uint16_t intercept_dr_read;
+    uint16_t intercept_dr_write;
+    uint32_t intercept_exceptions;
+    uint8_t v_tpr;
+
     /* KVM states, automatically cleared on reset */
     uint8_t nmi_injected;
     uint8_t nmi_pending;
@@ -888,7 +906,6 @@ typedef struct CPUX86State {
     CPU_COMMON
 
     /* Fields from here on are preserved across CPU reset. */
-    uint64_t pat;
 
     /* processor features (e.g. for CPUID insn) */
     uint32_t cpuid_level;
@@ -928,12 +945,7 @@ typedef struct CPUX86State {
     uint16_t fpus_vmstate;
     uint16_t fptag_vmstate;
     uint16_t fpregs_format_vmstate;
-
     uint64_t xstate_bv;
-    XMMReg ymmh_regs[CPU_NB_REGS];
-    BNDReg bnd_regs[4];
-    BNDCSReg bndcs_regs;
-    uint64_t msr_bndcfgs;
 
     uint64_t xcr0;
 
@@ -974,6 +986,7 @@ static inline void cpu_x86_load_seg_cache(CPUX86State *env,
     /* update the hidden flags */
     {
         if (seg_reg == R_CS) {
+            int cpl = selector & 3;
 #ifdef TARGET_X86_64
             if ((env->hflags & HF_LMA_MASK) && (flags & DESC_L_MASK)) {
                 /* long mode */
@@ -983,11 +996,19 @@ static inline void cpu_x86_load_seg_cache(CPUX86State *env,
 #endif
             {
                 /* legacy / compatibility case */
+                if (!(env->cr[0] & CR0_PE_MASK))
+                    cpl = 0;
+                else if (env->eflags & VM_MASK)
+                    cpl = 3;
                 new_hflags = (env->segs[R_CS].flags & DESC_B_MASK)
                     >> (DESC_B_SHIFT - HF_CS32_SHIFT);
                 env->hflags = (env->hflags & ~(HF_CS32_MASK | HF_CS64_MASK)) |
                     new_hflags;
             }
+#if HF_CPL_MASK != 3
+#error HF_CPL_MASK is hardcoded
+#endif
+            env->hflags = (env->hflags & ~HF_CPL_MASK) | cpl;
         }
         new_hflags = (env->segs[R_SS].flags & DESC_B_MASK)
             >> (DESC_B_SHIFT - HF_SS32_SHIFT);
@@ -1031,16 +1052,6 @@ int cpu_x86_get_descr_debug(CPUX86State *env, unsigned int selector,
                             target_ulong *base, unsigned int *limit,
                             unsigned int *flags);
 
-/* wrapper, just in case memory mappings must be changed */
-static inline void cpu_x86_set_cpl(CPUX86State *s, int cpl)
-{
-#if HF_CPL_MASK == 3
-    s->hflags = (s->hflags & ~HF_CPL_MASK) | cpl;
-#else
-#error HF_CPL_MASK is hardcoded
-#endif
-}
-
 /* op_helper.c */
 /* used for debug or cpu save/restore */
 void cpu_get_fp80(uint64_t *pmant, uint16_t *pexp, floatx80 f);
diff --git a/target-i386/helper.c b/target-i386/helper.c
index 372f0e3ecb..46d20e4b89 100644
--- a/target-i386/helper.c
+++ b/target-i386/helper.c
@@ -19,6 +19,7 @@
 
 #include "cpu.h"
 #include "sysemu/kvm.h"
+#include "kvm_i386.h"
 #ifndef CONFIG_USER_ONLY
 #include "sysemu/sysemu.h"
 #include "monitor/monitor.h"
@@ -1329,12 +1330,21 @@ void do_cpu_init(X86CPU *cpu)
 {
     CPUState *cs = CPU(cpu);
     CPUX86State *env = &cpu->env;
+    CPUX86State *save = g_new(CPUX86State, 1);
     int sipi = cs->interrupt_request & CPU_INTERRUPT_SIPI;
-    uint64_t pat = env->pat;
+
+    *save = *env;
 
     cpu_reset(cs);
     cs->interrupt_request = sipi;
-    env->pat = pat;
+    memcpy(&env->start_init_save, &save->start_init_save,
+           offsetof(CPUX86State, end_init_save) -
+           offsetof(CPUX86State, start_init_save));
+    g_free(save);
+
+    if (kvm_enabled()) {
+        kvm_arch_do_init_vcpu(cpu);
+    }
     apic_init_reset(cpu->apic_state);
 }
 
diff --git a/target-i386/kvm.c b/target-i386/kvm.c
index 4389959f61..0d894ef4aa 100644
--- a/target-i386/kvm.c
+++ b/target-i386/kvm.c
@@ -30,6 +30,8 @@
 #include "qemu/config-file.h"
 #include "hw/i386/pc.h"
 #include "hw/i386/apic.h"
+#include "hw/i386/apic_internal.h"
+#include "hw/i386/apic-msidef.h"
 #include "exec/ioport.h"
 #include <asm/hyperv.h>
 #include "hw/pci/pci.h"
@@ -130,14 +132,13 @@ static const struct kvm_para_features {
     { KVM_CAP_NOP_IO_DELAY, KVM_FEATURE_NOP_IO_DELAY },
     { KVM_CAP_PV_MMU, KVM_FEATURE_MMU_OP },
     { KVM_CAP_ASYNC_PF, KVM_FEATURE_ASYNC_PF },
-    { -1, -1 }
 };
 
 static int get_para_features(KVMState *s)
 {
     int i, features = 0;
 
-    for (i = 0; i < ARRAY_SIZE(para_features) - 1; i++) {
+    for (i = 0; i < ARRAY_SIZE(para_features); i++) {
         if (kvm_check_extension(s, para_features[i].cap)) {
             features |= (1 << para_features[i].feature);
         }
@@ -724,9 +725,8 @@ int kvm_arch_init_vcpu(CPUState *cs)
     return 0;
 }
 
-void kvm_arch_reset_vcpu(CPUState *cs)
+void kvm_arch_reset_vcpu(X86CPU *cpu)
 {
-    X86CPU *cpu = X86_CPU(cs);
     CPUX86State *env = &cpu->env;
 
     env->exception_injected = -1;
@@ -740,6 +740,16 @@ void kvm_arch_reset_vcpu(CPUState *cs)
     }
 }
 
+void kvm_arch_do_init_vcpu(X86CPU *cpu)
+{
+    CPUX86State *env = &cpu->env;
+
+    /* APs get directly into wait-for-SIPI state.  */
+    if (env->mp_state == KVM_MP_STATE_UNINITIALIZED) {
+        env->mp_state = KVM_MP_STATE_INIT_RECEIVED;
+    }
+}
+
 static int kvm_get_supported_msrs(KVMState *s)
 {
     static int kvm_supported_msrs;
@@ -2005,14 +2015,15 @@ void kvm_arch_pre_run(CPUState *cpu, struct kvm_run *run)
         }
     }
 
-    if (!kvm_irqchip_in_kernel()) {
-        /* Force the VCPU out of its inner loop to process any INIT requests
-         * or pending TPR access reports. */
-        if (cpu->interrupt_request &
-            (CPU_INTERRUPT_INIT | CPU_INTERRUPT_TPR)) {
-            cpu->exit_request = 1;
-        }
+    /* Force the VCPU out of its inner loop to process any INIT requests
+     * or (for userspace APIC, but it is cheap to combine the checks here)
+     * pending TPR access reports.
+     */
+    if (cpu->interrupt_request & (CPU_INTERRUPT_INIT | CPU_INTERRUPT_TPR)) {
+        cpu->exit_request = 1;
+    }
 
+    if (!kvm_irqchip_in_kernel()) {
         /* Try to inject an interrupt if the guest can accept it */
         if (run->ready_for_interrupt_injection &&
             (cpu->interrupt_request & CPU_INTERRUPT_HARD) &&
@@ -2092,6 +2103,11 @@ int kvm_arch_process_async_events(CPUState *cs)
         }
     }
 
+    if (cs->interrupt_request & CPU_INTERRUPT_INIT) {
+        kvm_cpu_synchronize_state(cs);
+        do_cpu_init(cpu);
+    }
+
     if (kvm_irqchip_in_kernel()) {
         return 0;
     }
@@ -2105,10 +2121,6 @@ int kvm_arch_process_async_events(CPUState *cs)
         (cs->interrupt_request & CPU_INTERRUPT_NMI)) {
         cs->halted = 0;
     }
-    if (cs->interrupt_request & CPU_INTERRUPT_INIT) {
-        kvm_cpu_synchronize_state(cs);
-        do_cpu_init(cpu);
-    }
     if (cs->interrupt_request & CPU_INTERRUPT_SIPI) {
         kvm_cpu_synchronize_state(cs);
         do_cpu_sipi(cpu);
diff --git a/target-i386/kvm_i386.h b/target-i386/kvm_i386.h
index 4392ab4359..cac30fd381 100644
--- a/target-i386/kvm_i386.h
+++ b/target-i386/kvm_i386.h
@@ -14,6 +14,8 @@
 #include "sysemu/kvm.h"
 
 bool kvm_allows_irq0_override(void);
+void kvm_arch_reset_vcpu(X86CPU *cs);
+void kvm_arch_do_init_vcpu(X86CPU *cs);
 
 int kvm_device_pci_assign(KVMState *s, PCIHostDeviceAddress *dev_addr,
                           uint32_t flags, uint32_t *dev_id);
diff --git a/target-i386/seg_helper.c b/target-i386/seg_helper.c
index 8c3f92c22b..3cf862ee60 100644
--- a/target-i386/seg_helper.c
+++ b/target-i386/seg_helper.c
@@ -409,11 +409,7 @@ static void switch_tss(CPUX86State *env, int tss_selector,
         for (i = 0; i < 6; i++) {
             load_seg_vm(env, i, new_segs[i]);
         }
-        /* in vm86, CPL is always 3 */
-        cpu_x86_set_cpl(env, 3);
     } else {
-        /* CPL is set the RPL of CS */
-        cpu_x86_set_cpl(env, new_segs[R_CS] & 3);
         /* first just selectors as the rest may trigger exceptions */
         for (i = 0; i < 6; i++) {
             cpu_x86_load_seg_cache(env, i, new_segs[i], 0, 0, 0);
@@ -739,6 +735,12 @@ static void do_interrupt_protected(CPUX86State *env, int intno, int is_int,
         }
     }
 
+    /* interrupt gate clear IF mask */
+    if ((type & 1) == 0) {
+        env->eflags &= ~IF_MASK;
+    }
+    env->eflags &= ~(TF_MASK | VM_MASK | RF_MASK | NT_MASK);
+
     if (new_stack) {
         if (env->eflags & VM_MASK) {
             cpu_x86_load_seg_cache(env, R_ES, 0, 0, 0, 0);
@@ -757,14 +759,7 @@ static void do_interrupt_protected(CPUX86State *env, int intno, int is_int,
                    get_seg_base(e1, e2),
                    get_seg_limit(e1, e2),
                    e2);
-    cpu_x86_set_cpl(env, dpl);
     env->eip = offset;
-
-    /* interrupt gate clear IF mask */
-    if ((type & 1) == 0) {
-        env->eflags &= ~IF_MASK;
-    }
-    env->eflags &= ~(TF_MASK | VM_MASK | RF_MASK | NT_MASK);
 }
 
 #ifdef TARGET_X86_64
@@ -911,6 +906,12 @@ static void do_interrupt64(CPUX86State *env, int intno, int is_int,
         PUSHQ(esp, error_code);
     }
 
+    /* interrupt gate clear IF mask */
+    if ((type & 1) == 0) {
+        env->eflags &= ~IF_MASK;
+    }
+    env->eflags &= ~(TF_MASK | VM_MASK | RF_MASK | NT_MASK);
+
     if (new_stack) {
         ss = 0 | dpl;
         cpu_x86_load_seg_cache(env, R_SS, ss, 0, 0, 0);
@@ -922,14 +923,7 @@ static void do_interrupt64(CPUX86State *env, int intno, int is_int,
                    get_seg_base(e1, e2),
                    get_seg_limit(e1, e2),
                    e2);
-    cpu_x86_set_cpl(env, dpl);
     env->eip = offset;
-
-    /* interrupt gate clear IF mask */
-    if ((type & 1) == 0) {
-        env->eflags &= ~IF_MASK;
-    }
-    env->eflags &= ~(TF_MASK | VM_MASK | RF_MASK | NT_MASK);
 }
 #endif
 
@@ -960,7 +954,8 @@ void helper_syscall(CPUX86State *env, int next_eip_addend)
 
         code64 = env->hflags & HF_CS64_MASK;
 
-        cpu_x86_set_cpl(env, 0);
+        env->eflags &= ~env->fmask;
+        cpu_load_eflags(env, env->eflags, 0);
         cpu_x86_load_seg_cache(env, R_CS, selector & 0xfffc,
                            0, 0xffffffff,
                                DESC_G_MASK | DESC_P_MASK |
@@ -972,8 +967,6 @@ void helper_syscall(CPUX86State *env, int next_eip_addend)
                                DESC_G_MASK | DESC_B_MASK | DESC_P_MASK |
                                DESC_S_MASK |
                                DESC_W_MASK | DESC_A_MASK);
-        env->eflags &= ~env->fmask;
-        cpu_load_eflags(env, env->eflags, 0);
         if (code64) {
             env->eip = env->lstar;
         } else {
@@ -982,7 +975,7 @@ void helper_syscall(CPUX86State *env, int next_eip_addend)
     } else {
         env->regs[R_ECX] = (uint32_t)(env->eip + next_eip_addend);
 
-        cpu_x86_set_cpl(env, 0);
+        env->eflags &= ~(IF_MASK | RF_MASK | VM_MASK);
         cpu_x86_load_seg_cache(env, R_CS, selector & 0xfffc,
                            0, 0xffffffff,
                                DESC_G_MASK | DESC_B_MASK | DESC_P_MASK |
@@ -993,7 +986,6 @@ void helper_syscall(CPUX86State *env, int next_eip_addend)
                                DESC_G_MASK | DESC_B_MASK | DESC_P_MASK |
                                DESC_S_MASK |
                                DESC_W_MASK | DESC_A_MASK);
-        env->eflags &= ~(IF_MASK | RF_MASK | VM_MASK);
         env->eip = (uint32_t)env->star;
     }
 }
@@ -1014,6 +1006,9 @@ void helper_sysret(CPUX86State *env, int dflag)
     }
     selector = (env->star >> 48) & 0xffff;
     if (env->hflags & HF_LMA_MASK) {
+        cpu_load_eflags(env, (uint32_t)(env->regs[11]), TF_MASK | AC_MASK
+                        | ID_MASK | IF_MASK | IOPL_MASK | VM_MASK | RF_MASK |
+                        NT_MASK);
         if (dflag == 2) {
             cpu_x86_load_seg_cache(env, R_CS, (selector + 16) | 3,
                                    0, 0xffffffff,
@@ -1035,11 +1030,8 @@ void helper_sysret(CPUX86State *env, int dflag)
                                DESC_G_MASK | DESC_B_MASK | DESC_P_MASK |
                                DESC_S_MASK | (3 << DESC_DPL_SHIFT) |
                                DESC_W_MASK | DESC_A_MASK);
-        cpu_load_eflags(env, (uint32_t)(env->regs[11]), TF_MASK | AC_MASK
-                        | ID_MASK | IF_MASK | IOPL_MASK | VM_MASK | RF_MASK |
-                        NT_MASK);
-        cpu_x86_set_cpl(env, 3);
     } else {
+        env->eflags |= IF_MASK;
         cpu_x86_load_seg_cache(env, R_CS, selector | 3,
                                0, 0xffffffff,
                                DESC_G_MASK | DESC_B_MASK | DESC_P_MASK |
@@ -1051,8 +1043,6 @@ void helper_sysret(CPUX86State *env, int dflag)
                                DESC_G_MASK | DESC_B_MASK | DESC_P_MASK |
                                DESC_S_MASK | (3 << DESC_DPL_SHIFT) |
                                DESC_W_MASK | DESC_A_MASK);
-        env->eflags |= IF_MASK;
-        cpu_x86_set_cpl(env, 3);
     }
 }
 #endif
@@ -1905,7 +1895,6 @@ void helper_lcall_protected(CPUX86State *env, int new_cs, target_ulong new_eip,
                        get_seg_base(e1, e2),
                        get_seg_limit(e1, e2),
                        e2);
-        cpu_x86_set_cpl(env, dpl);
         SET_ESP(sp, sp_mask);
         env->eip = offset;
     }
@@ -2134,7 +2123,6 @@ static inline void helper_ret_protected(CPUX86State *env, int shift,
                        get_seg_base(e1, e2),
                        get_seg_limit(e1, e2),
                        e2);
-        cpu_x86_set_cpl(env, rpl);
         sp = new_esp;
 #ifdef TARGET_X86_64
         if (env->hflags & HF_CS64_MASK) {
@@ -2185,7 +2173,6 @@ static inline void helper_ret_protected(CPUX86State *env, int shift,
                     IF_MASK | IOPL_MASK | VM_MASK | NT_MASK | VIF_MASK |
                     VIP_MASK);
     load_seg_vm(env, R_CS, new_cs & 0xffff);
-    cpu_x86_set_cpl(env, 3);
     load_seg_vm(env, R_SS, new_ss & 0xffff);
     load_seg_vm(env, R_ES, new_es & 0xffff);
     load_seg_vm(env, R_DS, new_ds & 0xffff);
@@ -2238,7 +2225,6 @@ void helper_sysenter(CPUX86State *env)
         raise_exception_err(env, EXCP0D_GPF, 0);
     }
     env->eflags &= ~(VM_MASK | IF_MASK | RF_MASK);
-    cpu_x86_set_cpl(env, 0);
 
 #ifdef TARGET_X86_64
     if (env->hflags & HF_LMA_MASK) {
@@ -2274,7 +2260,6 @@ void helper_sysexit(CPUX86State *env, int dflag)
     if (env->sysenter_cs == 0 || cpl != 0) {
         raise_exception_err(env, EXCP0D_GPF, 0);
     }
-    cpu_x86_set_cpl(env, 3);
 #ifdef TARGET_X86_64
     if (dflag == 2) {
         cpu_x86_load_seg_cache(env, R_CS, ((env->sysenter_cs + 32) & 0xfffc) |
diff --git a/target-i386/smm_helper.c b/target-i386/smm_helper.c
index 35901c9e89..4841d53b24 100644
--- a/target-i386/smm_helper.c
+++ b/target-i386/smm_helper.c
@@ -163,6 +163,13 @@ void do_smm_enter(X86CPU *cpu)
     cpu_load_eflags(env, 0, ~(CC_O | CC_S | CC_Z | CC_A | CC_P | CC_C |
                               DF_MASK));
     env->eip = 0x00008000;
+    cpu_x86_update_cr0(env,
+                       env->cr[0] & ~(CR0_PE_MASK | CR0_EM_MASK | CR0_TS_MASK |
+                                      CR0_PG_MASK));
+    cpu_x86_update_cr4(env, 0);
+    env->dr[7] = 0x00000400;
+    CC_OP = CC_OP_EFLAGS;
+
     cpu_x86_load_seg_cache(env, R_CS, (env->smbase >> 4) & 0xffff, env->smbase,
                            0xffffffff, 0);
     cpu_x86_load_seg_cache(env, R_DS, 0, 0, 0xffffffff, 0);
@@ -170,13 +177,6 @@ void do_smm_enter(X86CPU *cpu)
     cpu_x86_load_seg_cache(env, R_SS, 0, 0, 0xffffffff, 0);
     cpu_x86_load_seg_cache(env, R_FS, 0, 0, 0xffffffff, 0);
     cpu_x86_load_seg_cache(env, R_GS, 0, 0, 0xffffffff, 0);
-
-    cpu_x86_update_cr0(env,
-                       env->cr[0] & ~(CR0_PE_MASK | CR0_EM_MASK | CR0_TS_MASK |
-                                      CR0_PG_MASK));
-    cpu_x86_update_cr4(env, 0);
-    env->dr[7] = 0x00000400;
-    CC_OP = CC_OP_EFLAGS;
 }
 
 void helper_rsm(CPUX86State *env)
@@ -191,16 +191,6 @@ void helper_rsm(CPUX86State *env)
 #ifdef TARGET_X86_64
     cpu_load_efer(env, ldq_phys(cs->as, sm_state + 0x7ed0));
 
-    for (i = 0; i < 6; i++) {
-        offset = 0x7e00 + i * 16;
-        cpu_x86_load_seg_cache(env, i,
-                               lduw_phys(cs->as, sm_state + offset),
-                               ldq_phys(cs->as, sm_state + offset + 8),
-                               ldl_phys(cs->as, sm_state + offset + 4),
-                               (lduw_phys(cs->as, sm_state + offset + 2) &
-                                0xf0ff) << 8);
-    }
-
     env->gdt.base = ldq_phys(cs->as, sm_state + 0x7e68);
     env->gdt.limit = ldl_phys(cs->as, sm_state + 0x7e64);
 
@@ -238,6 +228,16 @@ void helper_rsm(CPUX86State *env)
     cpu_x86_update_cr3(env, ldl_phys(cs->as, sm_state + 0x7f50));
     cpu_x86_update_cr0(env, ldl_phys(cs->as, sm_state + 0x7f58));
 
+    for (i = 0; i < 6; i++) {
+        offset = 0x7e00 + i * 16;
+        cpu_x86_load_seg_cache(env, i,
+                               lduw_phys(cs->as, sm_state + offset),
+                               ldq_phys(cs->as, sm_state + offset + 8),
+                               ldl_phys(cs->as, sm_state + offset + 4),
+                               (lduw_phys(cs->as, sm_state + offset + 2) &
+                                0xf0ff) << 8);
+    }
+
     val = ldl_phys(cs->as, sm_state + 0x7efc); /* revision ID */
     if (val & 0x20000) {
         env->smbase = ldl_phys(cs->as, sm_state + 0x7f00) & ~0x7fff;
diff --git a/target-i386/svm_helper.c b/target-i386/svm_helper.c
index aa17ecdece..846eaa5918 100644
--- a/target-i386/svm_helper.c
+++ b/target-i386/svm_helper.c
@@ -282,9 +282,6 @@ void helper_vmrun(CPUX86State *env, int aflag, int next_eip_addend)
                           env->vm_vmcb + offsetof(struct vmcb, save.dr7));
     env->dr[6] = ldq_phys(cs->as,
                           env->vm_vmcb + offsetof(struct vmcb, save.dr6));
-    cpu_x86_set_cpl(env, ldub_phys(cs->as,
-                                   env->vm_vmcb + offsetof(struct vmcb,
-                                                           save.cpl)));
 
     /* FIXME: guest state consistency checks */
 
@@ -703,7 +700,8 @@ void helper_vmexit(CPUX86State *env, uint32_t exit_code, uint64_t exit_info_1)
     cpu_load_eflags(env, ldq_phys(cs->as,
                                   env->vm_hsave + offsetof(struct vmcb,
                                                            save.rflags)),
-                    ~(CC_O | CC_S | CC_Z | CC_A | CC_P | CC_C | DF_MASK));
+                    ~(CC_O | CC_S | CC_Z | CC_A | CC_P | CC_C | DF_MASK |
+                      VM_MASK));
     CC_OP = CC_OP_EFLAGS;
 
     svm_load_seg_cache(env, env->vm_hsave + offsetof(struct vmcb, save.es),
@@ -728,7 +726,6 @@ void helper_vmexit(CPUX86State *env, uint32_t exit_code, uint64_t exit_info_1)
                           env->vm_hsave + offsetof(struct vmcb, save.dr7));
 
     /* other setups */
-    cpu_x86_set_cpl(env, 0);
     stq_phys(cs->as, env->vm_vmcb + offsetof(struct vmcb, control.exit_code),
              exit_code);
     stq_phys(cs->as, env->vm_vmcb + offsetof(struct vmcb, control.exit_info_1),
@@ -756,10 +753,6 @@ void helper_vmexit(CPUX86State *env, uint32_t exit_code, uint64_t exit_info_1)
        from the page table indicated the host's CR3. If the PDPEs contain
        illegal state, the processor causes a shutdown. */
 
-    /* Forces CR0.PE = 1, RFLAGS.VM = 0. */
-    env->cr[0] |= CR0_PE_MASK;
-    env->eflags &= ~VM_MASK;
-
     /* Disables all breakpoints in the host DR7 register. */
 
     /* Checks the reloaded host state for consistency. */
diff --git a/target-i386/translate.c b/target-i386/translate.c
index 02625e31c2..032b0fdffc 100644
--- a/target-i386/translate.c
+++ b/target-i386/translate.c
@@ -6708,41 +6708,63 @@ static target_ulong disas_insn(CPUX86State *env, DisasContext *s,
         }
     bt_op:
         tcg_gen_andi_tl(cpu_T[1], cpu_T[1], (1 << (3 + ot)) - 1);
+        tcg_gen_shr_tl(cpu_tmp4, cpu_T[0], cpu_T[1]);
         switch(op) {
         case 0:
-            tcg_gen_shr_tl(cpu_cc_src, cpu_T[0], cpu_T[1]);
-            tcg_gen_movi_tl(cpu_cc_dst, 0);
             break;
         case 1:
-            tcg_gen_shr_tl(cpu_tmp4, cpu_T[0], cpu_T[1]);
             tcg_gen_movi_tl(cpu_tmp0, 1);
             tcg_gen_shl_tl(cpu_tmp0, cpu_tmp0, cpu_T[1]);
             tcg_gen_or_tl(cpu_T[0], cpu_T[0], cpu_tmp0);
             break;
         case 2:
-            tcg_gen_shr_tl(cpu_tmp4, cpu_T[0], cpu_T[1]);
             tcg_gen_movi_tl(cpu_tmp0, 1);
             tcg_gen_shl_tl(cpu_tmp0, cpu_tmp0, cpu_T[1]);
-            tcg_gen_not_tl(cpu_tmp0, cpu_tmp0);
-            tcg_gen_and_tl(cpu_T[0], cpu_T[0], cpu_tmp0);
+            tcg_gen_andc_tl(cpu_T[0], cpu_T[0], cpu_tmp0);
             break;
         default:
         case 3:
-            tcg_gen_shr_tl(cpu_tmp4, cpu_T[0], cpu_T[1]);
             tcg_gen_movi_tl(cpu_tmp0, 1);
             tcg_gen_shl_tl(cpu_tmp0, cpu_tmp0, cpu_T[1]);
             tcg_gen_xor_tl(cpu_T[0], cpu_T[0], cpu_tmp0);
             break;
         }
-        set_cc_op(s, CC_OP_SARB + ot);
         if (op != 0) {
             if (mod != 3) {
                 gen_op_st_v(s, ot, cpu_T[0], cpu_A0);
             } else {
                 gen_op_mov_reg_v(ot, rm, cpu_T[0]);
             }
+        }
+
+        /* Delay all CC updates until after the store above.  Note that
+           C is the result of the test, Z is unchanged, and the others
+           are all undefined.  */
+        switch (s->cc_op) {
+        case CC_OP_MULB ... CC_OP_MULQ:
+        case CC_OP_ADDB ... CC_OP_ADDQ:
+        case CC_OP_ADCB ... CC_OP_ADCQ:
+        case CC_OP_SUBB ... CC_OP_SUBQ:
+        case CC_OP_SBBB ... CC_OP_SBBQ:
+        case CC_OP_LOGICB ... CC_OP_LOGICQ:
+        case CC_OP_INCB ... CC_OP_INCQ:
+        case CC_OP_DECB ... CC_OP_DECQ:
+        case CC_OP_SHLB ... CC_OP_SHLQ:
+        case CC_OP_SARB ... CC_OP_SARQ:
+        case CC_OP_BMILGB ... CC_OP_BMILGQ:
+            /* Z was going to be computed from the non-zero status of CC_DST.
+               We can get that same Z value (and the new C value) by leaving
+               CC_DST alone, setting CC_SRC, and using a CC_OP_SAR of the
+               same width.  */
             tcg_gen_mov_tl(cpu_cc_src, cpu_tmp4);
-            tcg_gen_movi_tl(cpu_cc_dst, 0);
+            set_cc_op(s, ((s->cc_op - CC_OP_MULB) & 3) + CC_OP_SARB);
+            break;
+        default:
+            /* Otherwise, generate EFLAGS and replace the C bit.  */
+            gen_compute_eflags(s);
+            tcg_gen_deposit_tl(cpu_cc_src, cpu_cc_src, cpu_tmp4,
+                               ctz32(CC_C), 1);
+            break;
         }
         break;
     case 0x1bc: /* bsf / tzcnt */