diff --git a/Documentation/virtual/kvm/cpuid.txt b/Documentation/virtual/kvm/cpuid.txt
index 8820685..83afe65 100644
--- a/Documentation/virtual/kvm/cpuid.txt
+++ b/Documentation/virtual/kvm/cpuid.txt
@@ -10,11 +10,15 @@
 KVM cpuid functions are:
 
 function: KVM_CPUID_SIGNATURE (0x40000000)
-returns : eax = 0,
+returns : eax = 0x40000001,
           ebx = 0x4b4d564b,
           ecx = 0x564b4d56,
           edx = 0x4d.
 Note that this value in ebx, ecx and edx corresponds to the string "KVMKVMKVM".
+The value in eax corresponds to the maximum cpuid function present in this leaf,
+and will be updated if more functions are added in the future.
+Note also that old hosts set eax value to 0x0. This should
+be interpreted as if the value was 0x40000001.
 This function queries the presence of KVM cpuid leafs.
 
 
diff --git a/arch/x86/include/asm/kvm_para.h b/arch/x86/include/asm/kvm_para.h
index 99c4bbe..a7a7a94 100644
--- a/arch/x86/include/asm/kvm_para.h
+++ b/arch/x86/include/asm/kvm_para.h
@@ -178,14 +178,16 @@
 	unsigned int eax, ebx, ecx, edx;
 	char signature[13];
 
-	cpuid(KVM_CPUID_SIGNATURE, &eax, &ebx, &ecx, &edx);
-	memcpy(signature + 0, &ebx, 4);
-	memcpy(signature + 4, &ecx, 4);
-	memcpy(signature + 8, &edx, 4);
-	signature[12] = 0;
+	if (cpu_has_hypervisor) {
+		cpuid(KVM_CPUID_SIGNATURE, &eax, &ebx, &ecx, &edx);
+		memcpy(signature + 0, &ebx, 4);
+		memcpy(signature + 4, &ecx, 4);
+		memcpy(signature + 8, &edx, 4);
+		signature[12] = 0;
 
-	if (strcmp(signature, "KVMKVMKVM") == 0)
-		return 1;
+		if (strcmp(signature, "KVMKVMKVM") == 0)
+			return 1;
+	}
 
 	return 0;
 }
diff --git a/arch/x86/kvm/cpuid.c b/arch/x86/kvm/cpuid.c
index c2134b8..7df1c6d 100644
--- a/arch/x86/kvm/cpuid.c
+++ b/arch/x86/kvm/cpuid.c
@@ -398,7 +398,7 @@
 	case KVM_CPUID_SIGNATURE: {
 		char signature[12] = "KVMKVMKVM\0\0";
 		u32 *sigptr = (u32 *)signature;
-		entry->eax = 0;
+		entry->eax = KVM_CPUID_FEATURES;
 		entry->ebx = sigptr[0];
 		entry->ecx = sigptr[1];
 		entry->edx = sigptr[2];
diff --git a/arch/x86/kvm/emulate.c b/arch/x86/kvm/emulate.c
index 0d151e2..7fd2576 100644
--- a/arch/x86/kvm/emulate.c
+++ b/arch/x86/kvm/emulate.c
@@ -972,7 +972,6 @@
 		ctxt->modrm_rm = base_reg = (ctxt->rex_prefix & 1) << 3; /* REG.B */
 	}
 
-	ctxt->modrm = insn_fetch(u8, ctxt);
 	ctxt->modrm_mod |= (ctxt->modrm & 0xc0) >> 6;
 	ctxt->modrm_reg |= (ctxt->modrm & 0x38) >> 3;
 	ctxt->modrm_rm |= (ctxt->modrm & 0x07);
@@ -3359,8 +3358,8 @@
 		      .check_perm = (_p) }
 #define N    D(0)
 #define EXT(_f, _e) { .flags = ((_f) | RMExt), .u.group = (_e) }
-#define G(_f, _g) { .flags = ((_f) | Group), .u.group = (_g) }
-#define GD(_f, _g) { .flags = ((_f) | GroupDual), .u.gdual = (_g) }
+#define G(_f, _g) { .flags = ((_f) | Group | ModRM), .u.group = (_g) }
+#define GD(_f, _g) { .flags = ((_f) | GroupDual | ModRM), .u.gdual = (_g) }
 #define I(_f, _e) { .flags = (_f), .u.execute = (_e) }
 #define II(_f, _e, _i) \
 	{ .flags = (_f), .u.execute = (_e), .intercept = x86_intercept_##_i }
@@ -3380,25 +3379,25 @@
 		I2bv(((_f) & ~Lock) | DstAcc | SrcImm, _e)
 
 static struct opcode group7_rm1[] = {
-	DI(SrcNone | ModRM | Priv, monitor),
-	DI(SrcNone | ModRM | Priv, mwait),
+	DI(SrcNone | Priv, monitor),
+	DI(SrcNone | Priv, mwait),
 	N, N, N, N, N, N,
 };
 
 static struct opcode group7_rm3[] = {
-	DIP(SrcNone | ModRM | Prot | Priv, vmrun,   check_svme_pa),
-	II(SrcNone | ModRM | Prot | VendorSpecific, em_vmmcall, vmmcall),
-	DIP(SrcNone | ModRM | Prot | Priv, vmload,  check_svme_pa),
-	DIP(SrcNone | ModRM | Prot | Priv, vmsave,  check_svme_pa),
-	DIP(SrcNone | ModRM | Prot | Priv, stgi,    check_svme),
-	DIP(SrcNone | ModRM | Prot | Priv, clgi,    check_svme),
-	DIP(SrcNone | ModRM | Prot | Priv, skinit,  check_svme),
-	DIP(SrcNone | ModRM | Prot | Priv, invlpga, check_svme),
+	DIP(SrcNone | Prot | Priv,		vmrun,		check_svme_pa),
+	II(SrcNone  | Prot | VendorSpecific,	em_vmmcall,	vmmcall),
+	DIP(SrcNone | Prot | Priv,		vmload,		check_svme_pa),
+	DIP(SrcNone | Prot | Priv,		vmsave,		check_svme_pa),
+	DIP(SrcNone | Prot | Priv,		stgi,		check_svme),
+	DIP(SrcNone | Prot | Priv,		clgi,		check_svme),
+	DIP(SrcNone | Prot | Priv,		skinit,		check_svme),
+	DIP(SrcNone | Prot | Priv,		invlpga,	check_svme),
 };
 
 static struct opcode group7_rm7[] = {
 	N,
-	DIP(SrcNone | ModRM, rdtscp, check_rdtsc),
+	DIP(SrcNone, rdtscp, check_rdtsc),
 	N, N, N, N, N, N,
 };
 
@@ -3414,76 +3413,77 @@
 };
 
 static struct opcode group1A[] = {
-	I(DstMem | SrcNone | ModRM | Mov | Stack, em_pop), N, N, N, N, N, N, N,
+	I(DstMem | SrcNone | Mov | Stack, em_pop), N, N, N, N, N, N, N,
 };
 
 static struct opcode group3[] = {
-	I(DstMem | SrcImm | ModRM, em_test),
-	I(DstMem | SrcImm | ModRM, em_test),
-	I(DstMem | SrcNone | ModRM | Lock, em_not),
-	I(DstMem | SrcNone | ModRM | Lock, em_neg),
-	I(SrcMem | ModRM, em_mul_ex),
-	I(SrcMem | ModRM, em_imul_ex),
-	I(SrcMem | ModRM, em_div_ex),
-	I(SrcMem | ModRM, em_idiv_ex),
+	I(DstMem | SrcImm, em_test),
+	I(DstMem | SrcImm, em_test),
+	I(DstMem | SrcNone | Lock, em_not),
+	I(DstMem | SrcNone | Lock, em_neg),
+	I(SrcMem, em_mul_ex),
+	I(SrcMem, em_imul_ex),
+	I(SrcMem, em_div_ex),
+	I(SrcMem, em_idiv_ex),
 };
 
 static struct opcode group4[] = {
-	I(ByteOp | DstMem | SrcNone | ModRM | Lock, em_grp45),
-	I(ByteOp | DstMem | SrcNone | ModRM | Lock, em_grp45),
+	I(ByteOp | DstMem | SrcNone | Lock, em_grp45),
+	I(ByteOp | DstMem | SrcNone | Lock, em_grp45),
 	N, N, N, N, N, N,
 };
 
 static struct opcode group5[] = {
-	I(DstMem | SrcNone | ModRM | Lock, em_grp45),
-	I(DstMem | SrcNone | ModRM | Lock, em_grp45),
-	I(SrcMem | ModRM | Stack, em_grp45),
-	I(SrcMemFAddr | ModRM | ImplicitOps | Stack, em_call_far),
-	I(SrcMem | ModRM | Stack, em_grp45),
-	I(SrcMemFAddr | ModRM | ImplicitOps, em_grp45),
-	I(SrcMem | ModRM | Stack, em_grp45), N,
+	I(DstMem | SrcNone | Lock,		em_grp45),
+	I(DstMem | SrcNone | Lock,		em_grp45),
+	I(SrcMem | Stack,			em_grp45),
+	I(SrcMemFAddr | ImplicitOps | Stack,	em_call_far),
+	I(SrcMem | Stack,			em_grp45),
+	I(SrcMemFAddr | ImplicitOps,		em_grp45),
+	I(SrcMem | Stack,			em_grp45), N,
 };
 
 static struct opcode group6[] = {
-	DI(ModRM | Prot,        sldt),
-	DI(ModRM | Prot,        str),
-	DI(ModRM | Prot | Priv, lldt),
-	DI(ModRM | Prot | Priv, ltr),
+	DI(Prot,	sldt),
+	DI(Prot,	str),
+	DI(Prot | Priv,	lldt),
+	DI(Prot | Priv,	ltr),
 	N, N, N, N,
 };
 
 static struct group_dual group7 = { {
-	DI(ModRM | Mov | DstMem | Priv, sgdt),
-	DI(ModRM | Mov | DstMem | Priv, sidt),
-	II(ModRM | SrcMem | Priv, em_lgdt, lgdt),
-	II(ModRM | SrcMem | Priv, em_lidt, lidt),
-	II(SrcNone | ModRM | DstMem | Mov, em_smsw, smsw), N,
-	II(SrcMem16 | ModRM | Mov | Priv, em_lmsw, lmsw),
-	II(SrcMem | ModRM | ByteOp | Priv | NoAccess, em_invlpg, invlpg),
+	DI(Mov | DstMem | Priv,			sgdt),
+	DI(Mov | DstMem | Priv,			sidt),
+	II(SrcMem | Priv,			em_lgdt, lgdt),
+	II(SrcMem | Priv,			em_lidt, lidt),
+	II(SrcNone | DstMem | Mov,		em_smsw, smsw), N,
+	II(SrcMem16 | Mov | Priv,		em_lmsw, lmsw),
+	II(SrcMem | ByteOp | Priv | NoAccess,	em_invlpg, invlpg),
 }, {
-	I(SrcNone | ModRM | Priv | VendorSpecific, em_vmcall),
+	I(SrcNone | Priv | VendorSpecific,	em_vmcall),
 	EXT(0, group7_rm1),
 	N, EXT(0, group7_rm3),
-	II(SrcNone | ModRM | DstMem | Mov, em_smsw, smsw), N,
-	II(SrcMem16 | ModRM | Mov | Priv, em_lmsw, lmsw), EXT(0, group7_rm7),
+	II(SrcNone | DstMem | Mov,		em_smsw, smsw), N,
+	II(SrcMem16 | Mov | Priv,		em_lmsw, lmsw),
+	EXT(0, group7_rm7),
 } };
 
 static struct opcode group8[] = {
 	N, N, N, N,
-	I(DstMem | SrcImmByte | ModRM, em_bt),
-	I(DstMem | SrcImmByte | ModRM | Lock | PageTable, em_bts),
-	I(DstMem | SrcImmByte | ModRM | Lock, em_btr),
-	I(DstMem | SrcImmByte | ModRM | Lock | PageTable, em_btc),
+	I(DstMem | SrcImmByte,				em_bt),
+	I(DstMem | SrcImmByte | Lock | PageTable,	em_bts),
+	I(DstMem | SrcImmByte | Lock,			em_btr),
+	I(DstMem | SrcImmByte | Lock | PageTable,	em_btc),
 };
 
 static struct group_dual group9 = { {
-	N, I(DstMem64 | ModRM | Lock | PageTable, em_cmpxchg8b), N, N, N, N, N, N,
+	N, I(DstMem64 | Lock | PageTable, em_cmpxchg8b), N, N, N, N, N, N,
 }, {
 	N, N, N, N, N, N, N, N,
 } };
 
 static struct opcode group11[] = {
-	I(DstMem | SrcImm | ModRM | Mov | PageTable, em_mov),
+	I(DstMem | SrcImm | Mov | PageTable, em_mov),
 	X7(D(Undefined)),
 };
 
@@ -3541,10 +3541,10 @@
 	/* 0x70 - 0x7F */
 	X16(D(SrcImmByte)),
 	/* 0x80 - 0x87 */
-	G(ByteOp | DstMem | SrcImm | ModRM | Group, group1),
-	G(DstMem | SrcImm | ModRM | Group, group1),
-	G(ByteOp | DstMem | SrcImm | ModRM | No64 | Group, group1),
-	G(DstMem | SrcImmByte | ModRM | Group, group1),
+	G(ByteOp | DstMem | SrcImm, group1),
+	G(DstMem | SrcImm, group1),
+	G(ByteOp | DstMem | SrcImm | No64, group1),
+	G(DstMem | SrcImmByte, group1),
 	I2bv(DstMem | SrcReg | ModRM, em_test),
 	I2bv(DstMem | SrcReg | ModRM | Lock | PageTable, em_xchg),
 	/* 0x88 - 0x8F */
@@ -3975,17 +3975,16 @@
 	}
 	ctxt->d = opcode.flags;
 
+	if (ctxt->d & ModRM)
+		ctxt->modrm = insn_fetch(u8, ctxt);
+
 	while (ctxt->d & GroupMask) {
 		switch (ctxt->d & GroupMask) {
 		case Group:
-			ctxt->modrm = insn_fetch(u8, ctxt);
-			--ctxt->_eip;
 			goffset = (ctxt->modrm >> 3) & 7;
 			opcode = opcode.u.group[goffset];
 			break;
 		case GroupDual:
-			ctxt->modrm = insn_fetch(u8, ctxt);
-			--ctxt->_eip;
 			goffset = (ctxt->modrm >> 3) & 7;
 			if ((ctxt->modrm >> 6) == 3)
 				opcode = opcode.u.gdual->mod3[goffset];