1 files changed, 220 insertions, 81 deletions

diff --git a/arch/x86/kvm/emulate.c b/arch/x86/kvm/emulate.c
index 03954f7900f5..749f9fa38254 100644
--- a/arch/x86/kvm/emulate.c
+++ b/arch/x86/kvm/emulate.c
@@ -504,11 +504,6 @@ static void rsp_increment(struct x86_emulate_ctxt *ctxt, int inc)
 	masked_increment(reg_rmw(ctxt, VCPU_REGS_RSP), stack_mask(ctxt), inc);
 }
 
-static inline void jmp_rel(struct x86_emulate_ctxt *ctxt, int rel)
-{
-	register_address_increment(ctxt, &ctxt->_eip, rel);
-}
-
 static u32 desc_limit_scaled(struct desc_struct *desc)
 {
 	u32 limit = get_desc_limit(desc);
@@ -527,6 +522,7 @@ static unsigned long seg_base(struct x86_emulate_ctxt *ctxt, int seg)
 static int emulate_exception(struct x86_emulate_ctxt *ctxt, int vec,
 			     u32 error, bool valid)
 {
+	WARN_ON(vec > 0x1f);
 	ctxt->exception.vector = vec;
 	ctxt->exception.error_code = error;
 	ctxt->exception.error_code_valid = valid;
@@ -568,6 +564,38 @@ static int emulate_nm(struct x86_emulate_ctxt *ctxt)
 	return emulate_exception(ctxt, NM_VECTOR, 0, false);
 }
 
+static inline int assign_eip_far(struct x86_emulate_ctxt *ctxt, ulong dst,
+			       int cs_l)
+{
+	switch (ctxt->op_bytes) {
+	case 2:
+		ctxt->_eip = (u16)dst;
+		break;
+	case 4:
+		ctxt->_eip = (u32)dst;
+		break;
+	case 8:
+		if ((cs_l && is_noncanonical_address(dst)) ||
+		    (!cs_l && (dst & ~(u32)-1)))
+			return emulate_gp(ctxt, 0);
+		ctxt->_eip = dst;
+		break;
+	default:
+		WARN(1, "unsupported eip assignment size\n");
+	}
+	return X86EMUL_CONTINUE;
+}
+
+static inline int assign_eip_near(struct x86_emulate_ctxt *ctxt, ulong dst)
+{
+	return assign_eip_far(ctxt, dst, ctxt->mode == X86EMUL_MODE_PROT64);
+}
+
+static inline int jmp_rel(struct x86_emulate_ctxt *ctxt, int rel)
+{
+	return assign_eip_near(ctxt, ctxt->_eip + rel);
+}
+
 static u16 get_segment_selector(struct x86_emulate_ctxt *ctxt, unsigned seg)
 {
 	u16 selector;
@@ -750,8 +778,10 @@ static int __do_insn_fetch_bytes(struct x86_emulate_ctxt *ctxt, int op_size)
 static __always_inline int do_insn_fetch_bytes(struct x86_emulate_ctxt *ctxt,
 					       unsigned size)
 {
-	if (unlikely(ctxt->fetch.end - ctxt->fetch.ptr < size))
-		return __do_insn_fetch_bytes(ctxt, size);
+	unsigned done_size = ctxt->fetch.end - ctxt->fetch.ptr;
+
+	if (unlikely(done_size < size))
+		return __do_insn_fetch_bytes(ctxt, size - done_size);
 	else
 		return X86EMUL_CONTINUE;
 }
@@ -1415,7 +1445,9 @@ static int write_segment_descriptor(struct x86_emulate_ctxt *ctxt,
 
 /* Does not support long mode */
 static int __load_segment_descriptor(struct x86_emulate_ctxt *ctxt,
-				     u16 selector, int seg, u8 cpl, bool in_task_switch)
+				     u16 selector, int seg, u8 cpl,
+				     bool in_task_switch,
+				     struct desc_struct *desc)
 {
 	struct desc_struct seg_desc, old_desc;
 	u8 dpl, rpl;
@@ -1468,7 +1500,7 @@ static int __load_segment_descriptor(struct x86_emulate_ctxt *ctxt,
 		return ret;
 
 	err_code = selector & 0xfffc;
-	err_vec = GP_VECTOR;
+	err_vec = in_task_switch ? TS_VECTOR : GP_VECTOR;
 
 	/* can't load system descriptor into segment selector */
 	if (seg <= VCPU_SREG_GS && !seg_desc.s)
@@ -1503,6 +1535,15 @@ static int __load_segment_descriptor(struct x86_emulate_ctxt *ctxt,
 			if (rpl > cpl || dpl != cpl)
 				goto exception;
 		}
+		/* in long-mode d/b must be clear if l is set */
+		if (seg_desc.d && seg_desc.l) {
+			u64 efer = 0;
+
+			ctxt->ops->get_msr(ctxt, MSR_EFER, &efer);
+			if (efer & EFER_LMA)
+				goto exception;
+		}
+
 		/* CS(RPL) <- CPL */
 		selector = (selector & 0xfffc) | cpl;
 		break;
@@ -1547,17 +1588,18 @@ static int __load_segment_descriptor(struct x86_emulate_ctxt *ctxt,
 	}
 load:
 	ctxt->ops->set_segment(ctxt, selector, &seg_desc, base3, seg);
+	if (desc)
+		*desc = seg_desc;
 	return X86EMUL_CONTINUE;
 exception:
-	emulate_exception(ctxt, err_vec, err_code, true);
-	return X86EMUL_PROPAGATE_FAULT;
+	return emulate_exception(ctxt, err_vec, err_code, true);
 }
 
 static int load_segment_descriptor(struct x86_emulate_ctxt *ctxt,
 				   u16 selector, int seg)
 {
 	u8 cpl = ctxt->ops->cpl(ctxt);
-	return __load_segment_descriptor(ctxt, selector, seg, cpl, false);
+	return __load_segment_descriptor(ctxt, selector, seg, cpl, false, NULL);
 }
 
 static void write_register_operand(struct operand *op)
@@ -1951,17 +1993,31 @@ static int em_iret(struct x86_emulate_ctxt *ctxt)
 static int em_jmp_far(struct x86_emulate_ctxt *ctxt)
 {
 	int rc;
-	unsigned short sel;
+	unsigned short sel, old_sel;
+	struct desc_struct old_desc, new_desc;
+	const struct x86_emulate_ops *ops = ctxt->ops;
+	u8 cpl = ctxt->ops->cpl(ctxt);
+
+	/* Assignment of RIP may only fail in 64-bit mode */
+	if (ctxt->mode == X86EMUL_MODE_PROT64)
+		ops->get_segment(ctxt, &old_sel, &old_desc, NULL,
+				 VCPU_SREG_CS);
 
 	memcpy(&sel, ctxt->src.valptr + ctxt->op_bytes, 2);
 
-	rc = load_segment_descriptor(ctxt, sel, VCPU_SREG_CS);
+	rc = __load_segment_descriptor(ctxt, sel, VCPU_SREG_CS, cpl, false,
+				       &new_desc);
 	if (rc != X86EMUL_CONTINUE)
 		return rc;
 
-	ctxt->_eip = 0;
-	memcpy(&ctxt->_eip, ctxt->src.valptr, ctxt->op_bytes);
-	return X86EMUL_CONTINUE;
+	rc = assign_eip_far(ctxt, ctxt->src.val, new_desc.l);
+	if (rc != X86EMUL_CONTINUE) {
+		WARN_ON(!ctxt->mode != X86EMUL_MODE_PROT64);
+		/* assigning eip failed; restore the old cs */
+		ops->set_segment(ctxt, old_sel, &old_desc, 0, VCPU_SREG_CS);
+		return rc;
+	}
+	return rc;
 }
 
 static int em_grp45(struct x86_emulate_ctxt *ctxt)
@@ -1972,13 +2028,15 @@ static int em_grp45(struct x86_emulate_ctxt *ctxt)
 	case 2: /* call near abs */ {
 		long int old_eip;
 		old_eip = ctxt->_eip;
-		ctxt->_eip = ctxt->src.val;
+		rc = assign_eip_near(ctxt, ctxt->src.val);
+		if (rc != X86EMUL_CONTINUE)
+			break;
 		ctxt->src.val = old_eip;
 		rc = em_push(ctxt);
 		break;
 	}
 	case 4: /* jmp abs */
-		ctxt->_eip = ctxt->src.val;
+		rc = assign_eip_near(ctxt, ctxt->src.val);
 		break;
 	case 5: /* jmp far */
 		rc = em_jmp_far(ctxt);
@@ -2013,30 +2071,47 @@ static int em_cmpxchg8b(struct x86_emulate_ctxt *ctxt)
 
 static int em_ret(struct x86_emulate_ctxt *ctxt)
 {
-	ctxt->dst.type = OP_REG;
-	ctxt->dst.addr.reg = &ctxt->_eip;
-	ctxt->dst.bytes = ctxt->op_bytes;
-	return em_pop(ctxt);
+	int rc;
+	unsigned long eip;
+
+	rc = emulate_pop(ctxt, &eip, ctxt->op_bytes);
+	if (rc != X86EMUL_CONTINUE)
+		return rc;
+
+	return assign_eip_near(ctxt, eip);
 }
 
 static int em_ret_far(struct x86_emulate_ctxt *ctxt)
 {
 	int rc;
-	unsigned long cs;
+	unsigned long eip, cs;
+	u16 old_cs;
 	int cpl = ctxt->ops->cpl(ctxt);
+	struct desc_struct old_desc, new_desc;
+	const struct x86_emulate_ops *ops = ctxt->ops;
+
+	if (ctxt->mode == X86EMUL_MODE_PROT64)
+		ops->get_segment(ctxt, &old_cs, &old_desc, NULL,
+				 VCPU_SREG_CS);
 
-	rc = emulate_pop(ctxt, &ctxt->_eip, ctxt->op_bytes);
+	rc = emulate_pop(ctxt, &eip, ctxt->op_bytes);
 	if (rc != X86EMUL_CONTINUE)
 		return rc;
-	if (ctxt->op_bytes == 4)
-		ctxt->_eip = (u32)ctxt->_eip;
 	rc = emulate_pop(ctxt, &cs, ctxt->op_bytes);
 	if (rc != X86EMUL_CONTINUE)
 		return rc;
 	/* Outer-privilege level return is not implemented */
 	if (ctxt->mode >= X86EMUL_MODE_PROT16 && (cs & 3) > cpl)
 		return X86EMUL_UNHANDLEABLE;
-	rc = load_segment_descriptor(ctxt, (u16)cs, VCPU_SREG_CS);
+	rc = __load_segment_descriptor(ctxt, (u16)cs, VCPU_SREG_CS, 0, false,
+				       &new_desc);
+	if (rc != X86EMUL_CONTINUE)
+		return rc;
+	rc = assign_eip_far(ctxt, eip, new_desc.l);
+	if (rc != X86EMUL_CONTINUE) {
+		WARN_ON(!ctxt->mode != X86EMUL_MODE_PROT64);
+		ops->set_segment(ctxt, old_cs, &old_desc, 0, VCPU_SREG_CS);
+	}
 	return rc;
 }
 
@@ -2297,7 +2372,7 @@ static int em_sysexit(struct x86_emulate_ctxt *ctxt)
 {
 	const struct x86_emulate_ops *ops = ctxt->ops;
 	struct desc_struct cs, ss;
-	u64 msr_data;
+	u64 msr_data, rcx, rdx;
 	int usermode;
 	u16 cs_sel = 0, ss_sel = 0;
 
@@ -2313,6 +2388,9 @@ static int em_sysexit(struct x86_emulate_ctxt *ctxt)
 	else
 		usermode = X86EMUL_MODE_PROT32;
 
+	rcx = reg_read(ctxt, VCPU_REGS_RCX);
+	rdx = reg_read(ctxt, VCPU_REGS_RDX);
+
 	cs.dpl = 3;
 	ss.dpl = 3;
 	ops->get_msr(ctxt, MSR_IA32_SYSENTER_CS, &msr_data);
@@ -2330,6 +2408,9 @@ static int em_sysexit(struct x86_emulate_ctxt *ctxt)
 		ss_sel = cs_sel + 8;
 		cs.d = 0;
 		cs.l = 1;
+		if (is_noncanonical_address(rcx) ||
+		    is_noncanonical_address(rdx))
+			return emulate_gp(ctxt, 0);
 		break;
 	}
 	cs_sel |= SELECTOR_RPL_MASK;
@@ -2338,8 +2419,8 @@ static int em_sysexit(struct x86_emulate_ctxt *ctxt)
 	ops->set_segment(ctxt, cs_sel, &cs, 0, VCPU_SREG_CS);
 	ops->set_segment(ctxt, ss_sel, &ss, 0, VCPU_SREG_SS);
 
-	ctxt->_eip = reg_read(ctxt, VCPU_REGS_RDX);
-	*reg_write(ctxt, VCPU_REGS_RSP) = reg_read(ctxt, VCPU_REGS_RCX);
+	ctxt->_eip = rdx;
+	*reg_write(ctxt, VCPU_REGS_RSP) = rcx;
 
 	return X86EMUL_CONTINUE;
 }
@@ -2457,19 +2538,24 @@ static int load_state_from_tss16(struct x86_emulate_ctxt *ctxt,
 	 * Now load segment descriptors. If fault happens at this stage
 	 * it is handled in a context of new task
 	 */
-	ret = __load_segment_descriptor(ctxt, tss->ldt, VCPU_SREG_LDTR, cpl, true);
+	ret = __load_segment_descriptor(ctxt, tss->ldt, VCPU_SREG_LDTR, cpl,
+					true, NULL);
 	if (ret != X86EMUL_CONTINUE)
 		return ret;
-	ret = __load_segment_descriptor(ctxt, tss->es, VCPU_SREG_ES, cpl, true);
+	ret = __load_segment_descriptor(ctxt, tss->es, VCPU_SREG_ES, cpl,
+					true, NULL);
 	if (ret != X86EMUL_CONTINUE)
 		return ret;
-	ret = __load_segment_descriptor(ctxt, tss->cs, VCPU_SREG_CS, cpl, true);
+	ret = __load_segment_descriptor(ctxt, tss->cs, VCPU_SREG_CS, cpl,
+					true, NULL);
 	if (ret != X86EMUL_CONTINUE)
 		return ret;
-	ret = __load_segment_descriptor(ctxt, tss->ss, VCPU_SREG_SS, cpl, true);
+	ret = __load_segment_descriptor(ctxt, tss->ss, VCPU_SREG_SS, cpl,
+					true, NULL);
 	if (ret != X86EMUL_CONTINUE)
 		return ret;
-	ret = __load_segment_descriptor(ctxt, tss->ds, VCPU_SREG_DS, cpl, true);
+	ret = __load_segment_descriptor(ctxt, tss->ds, VCPU_SREG_DS, cpl,
+					true, NULL);
 	if (ret != X86EMUL_CONTINUE)
 		return ret;
 
@@ -2594,25 +2680,32 @@ static int load_state_from_tss32(struct x86_emulate_ctxt *ctxt,
 	 * Now load segment descriptors. If fault happenes at this stage
 	 * it is handled in a context of new task
 	 */
-	ret = __load_segment_descriptor(ctxt, tss->ldt_selector, VCPU_SREG_LDTR, cpl, true);
+	ret = __load_segment_descriptor(ctxt, tss->ldt_selector, VCPU_SREG_LDTR,
+					cpl, true, NULL);
 	if (ret != X86EMUL_CONTINUE)
 		return ret;
-	ret = __load_segment_descriptor(ctxt, tss->es, VCPU_SREG_ES, cpl, true);
+	ret = __load_segment_descriptor(ctxt, tss->es, VCPU_SREG_ES, cpl,
+					true, NULL);
 	if (ret != X86EMUL_CONTINUE)
 		return ret;
-	ret = __load_segment_descriptor(ctxt, tss->cs, VCPU_SREG_CS, cpl, true);
+	ret = __load_segment_descriptor(ctxt, tss->cs, VCPU_SREG_CS, cpl,
+					true, NULL);
 	if (ret != X86EMUL_CONTINUE)
 		return ret;
-	ret = __load_segment_descriptor(ctxt, tss->ss, VCPU_SREG_SS, cpl, true);
+	ret = __load_segment_descriptor(ctxt, tss->ss, VCPU_SREG_SS, cpl,
+					true, NULL);
 	if (ret != X86EMUL_CONTINUE)
 		return ret;
-	ret = __load_segment_descriptor(ctxt, tss->ds, VCPU_SREG_DS, cpl, true);
+	ret = __load_segment_descriptor(ctxt, tss->ds, VCPU_SREG_DS, cpl,
+					true, NULL);
 	if (ret != X86EMUL_CONTINUE)
 		return ret;
-	ret = __load_segment_descriptor(ctxt, tss->fs, VCPU_SREG_FS, cpl, true);
+	ret = __load_segment_descriptor(ctxt, tss->fs, VCPU_SREG_FS, cpl,
+					true, NULL);
 	if (ret != X86EMUL_CONTINUE)
 		return ret;
-	ret = __load_segment_descriptor(ctxt, tss->gs, VCPU_SREG_GS, cpl, true);
+	ret = __load_segment_descriptor(ctxt, tss->gs, VCPU_SREG_GS, cpl,
+					true, NULL);
 	if (ret != X86EMUL_CONTINUE)
 		return ret;
 
@@ -2723,8 +2816,7 @@ static int emulator_do_task_switch(struct x86_emulate_ctxt *ctxt,
 	if (!next_tss_desc.p ||
 	    ((desc_limit < 0x67 && (next_tss_desc.type & 8)) ||
 	     desc_limit < 0x2b)) {
-		emulate_ts(ctxt, tss_selector & 0xfffc);
-		return X86EMUL_PROPAGATE_FAULT;
+		return emulate_ts(ctxt, tss_selector & 0xfffc);
 	}
 
 	if (reason == TASK_SWITCH_IRET || reason == TASK_SWITCH_JMP) {
@@ -2880,10 +2972,13 @@ static int em_aad(struct x86_emulate_ctxt *ctxt)
 
 static int em_call(struct x86_emulate_ctxt *ctxt)
 {
+	int rc;
 	long rel = ctxt->src.val;
 
 	ctxt->src.val = (unsigned long)ctxt->_eip;
-	jmp_rel(ctxt, rel);
+	rc = jmp_rel(ctxt, rel);
+	if (rc != X86EMUL_CONTINUE)
+		return rc;
 	return em_push(ctxt);
 }
 
@@ -2892,34 +2987,50 @@ static int em_call_far(struct x86_emulate_ctxt *ctxt)
 	u16 sel, old_cs;
 	ulong old_eip;
 	int rc;
+	struct desc_struct old_desc, new_desc;
+	const struct x86_emulate_ops *ops = ctxt->ops;
+	int cpl = ctxt->ops->cpl(ctxt);
 
-	old_cs = get_segment_selector(ctxt, VCPU_SREG_CS);
 	old_eip = ctxt->_eip;
+	ops->get_segment(ctxt, &old_cs, &old_desc, NULL, VCPU_SREG_CS);
 
 	memcpy(&sel, ctxt->src.valptr + ctxt->op_bytes, 2);
-	if (load_segment_descriptor(ctxt, sel, VCPU_SREG_CS))
+	rc = __load_segment_descriptor(ctxt, sel, VCPU_SREG_CS, cpl, false,
+				       &new_desc);
+	if (rc != X86EMUL_CONTINUE)
 		return X86EMUL_CONTINUE;
 
-	ctxt->_eip = 0;
-	memcpy(&ctxt->_eip, ctxt->src.valptr, ctxt->op_bytes);
+	rc = assign_eip_far(ctxt, ctxt->src.val, new_desc.l);
+	if (rc != X86EMUL_CONTINUE)
+		goto fail;
 
 	ctxt->src.val = old_cs;
 	rc = em_push(ctxt);
 	if (rc != X86EMUL_CONTINUE)
-		return rc;
+		goto fail;
 
 	ctxt->src.val = old_eip;
-	return em_push(ctxt);
+	rc = em_push(ctxt);
+	/* If we failed, we tainted the memory, but the very least we should
+	   restore cs */
+	if (rc != X86EMUL_CONTINUE)
+		goto fail;
+	return rc;
+fail:
+	ops->set_segment(ctxt, old_cs, &old_desc, 0, VCPU_SREG_CS);
+	return rc;
+
 }
 
 static int em_ret_near_imm(struct x86_emulate_ctxt *ctxt)
 {
 	int rc;
+	unsigned long eip;
 
-	ctxt->dst.type = OP_REG;
-	ctxt->dst.addr.reg = &ctxt->_eip;
-	ctxt->dst.bytes = ctxt->op_bytes;
-	rc = emulate_pop(ctxt, &ctxt->dst.val, ctxt->op_bytes);
+	rc = emulate_pop(ctxt, &eip, ctxt->op_bytes);
+	if (rc != X86EMUL_CONTINUE)
+		return rc;
+	rc = assign_eip_near(ctxt, eip);
 	if (rc != X86EMUL_CONTINUE)
 		return rc;
 	rsp_increment(ctxt, ctxt->src.val);
@@ -3016,7 +3127,7 @@ static int em_movbe(struct x86_emulate_ctxt *ctxt)
 		ctxt->dst.val = swab64(ctxt->src.val);
 		break;
 	default:
-		return X86EMUL_PROPAGATE_FAULT;
+		BUG();
 	}
 	return X86EMUL_CONTINUE;
 }
@@ -3140,12 +3251,8 @@ static int em_clts(struct x86_emulate_ctxt *ctxt)
 
 static int em_vmcall(struct x86_emulate_ctxt *ctxt)
 {
-	int rc;
+	int rc = ctxt->ops->fix_hypercall(ctxt);
 
-	if (ctxt->modrm_mod != 3 || ctxt->modrm_rm != 1)
-		return X86EMUL_UNHANDLEABLE;
-
-	rc = ctxt->ops->fix_hypercall(ctxt);
 	if (rc != X86EMUL_CONTINUE)
 		return rc;
 
@@ -3250,20 +3357,24 @@ static int em_lmsw(struct x86_emulate_ctxt *ctxt)
 
 static int em_loop(struct x86_emulate_ctxt *ctxt)
 {
+	int rc = X86EMUL_CONTINUE;
+
 	register_address_increment(ctxt, reg_rmw(ctxt, VCPU_REGS_RCX), -1);
 	if ((address_mask(ctxt, reg_read(ctxt, VCPU_REGS_RCX)) != 0) &&
 	    (ctxt->b == 0xe2 || test_cc(ctxt->b ^ 0x5, ctxt->eflags)))
-		jmp_rel(ctxt, ctxt->src.val);
+		rc = jmp_rel(ctxt, ctxt->src.val);
 
-	return X86EMUL_CONTINUE;
+	return rc;
 }
 
 static int em_jcxz(struct x86_emulate_ctxt *ctxt)
 {
+	int rc = X86EMUL_CONTINUE;
+
 	if (address_mask(ctxt, reg_read(ctxt, VCPU_REGS_RCX)) == 0)
-		jmp_rel(ctxt, ctxt->src.val);
+		rc = jmp_rel(ctxt, ctxt->src.val);
 
-	return X86EMUL_CONTINUE;
+	return rc;
 }
 
 static int em_in(struct x86_emulate_ctxt *ctxt)
@@ -3351,6 +3462,12 @@ static int em_bswap(struct x86_emulate_ctxt *ctxt)
 	return X86EMUL_CONTINUE;
 }
 
+static int em_clflush(struct x86_emulate_ctxt *ctxt)
+{
+	/* emulating clflush regardless of cpuid */
+	return X86EMUL_CONTINUE;
+}
+
 static bool valid_cr(int nr)
 {
 	switch (nr) {
@@ -3563,6 +3680,12 @@ static int check_perm_out(struct x86_emulate_ctxt *ctxt)
 		F2bv(((_f) | DstReg | SrcMem | ModRM) & ~Lock, _e),	\
 		F2bv(((_f) & ~Lock) | DstAcc | SrcImm, _e)
 
+static const struct opcode group7_rm0[] = {
+	N,
+	I(SrcNone | Priv | EmulateOnUD,	em_vmcall),
+	N, N, N, N, N, N,
+};
+
 static const struct opcode group7_rm1[] = {
 	DI(SrcNone | Priv, monitor),
 	DI(SrcNone | Priv, mwait),
@@ -3656,7 +3779,7 @@ static const struct group_dual group7 = { {
 	II(SrcMem16 | Mov | Priv,		em_lmsw, lmsw),
 	II(SrcMem | ByteOp | Priv | NoAccess,	em_invlpg, invlpg),
 }, {
-	I(SrcNone | Priv | EmulateOnUD,	em_vmcall),
+	EXT(0, group7_rm0),
 	EXT(0, group7_rm1),
 	N, EXT(0, group7_rm3),
 	II(SrcNone | DstMem | Mov,		em_smsw, smsw), N,
@@ -3683,18 +3806,32 @@ static const struct opcode group11[] = {
 	X7(D(Undefined)),
 };
 
+static const struct gprefix pfx_0f_ae_7 = {
+	I(SrcMem | ByteOp, em_clflush), N, N, N,
+};
+
+static const struct group_dual group15 = { {
+	N, N, N, N, N, N, N, GP(0, &pfx_0f_ae_7),
+}, {
+	N, N, N, N, N, N, N, N,
+} };
+
 static const struct gprefix pfx_0f_6f_0f_7f = {
 	I(Mmx, em_mov), I(Sse | Aligned, em_mov), N, I(Sse | Unaligned, em_mov),
 };
 
-static const struct gprefix pfx_vmovntpx = {
-	I(0, em_mov), N, N, N,
+static const struct gprefix pfx_0f_2b = {
+	I(0, em_mov), I(0, em_mov), N, N,
 };
 
 static const struct gprefix pfx_0f_28_0f_29 = {
 	I(Aligned, em_mov), I(Aligned, em_mov), N, N,
 };
 
+static const struct gprefix pfx_0f_e7 = {
+	N, I(Sse, em_mov), N, N,
+};
+
 static const struct escape escape_d9 = { {
 	N, N, N, N, N, N, N, I(DstMem, em_fnstcw),
 }, {
@@ -3887,10 +4024,11 @@ static const struct opcode twobyte_table[256] = {
 	N, I(ImplicitOps | EmulateOnUD, em_syscall),
 	II(ImplicitOps | Priv, em_clts, clts), N,
 	DI(ImplicitOps | Priv, invd), DI(ImplicitOps | Priv, wbinvd), N, N,
-	N, D(ImplicitOps | ModRM), N, N,
+	N, D(ImplicitOps | ModRM | SrcMem | NoAccess), N, N,
 	/* 0x10 - 0x1F */
 	N, N, N, N, N, N, N, N,
-	D(ImplicitOps | ModRM), N, N, N, N, N, N, D(ImplicitOps | ModRM),
+	D(ImplicitOps | ModRM | SrcMem | NoAccess),
+	N, N, N, N, N, N, D(ImplicitOps | ModRM | SrcMem | NoAccess),
 	/* 0x20 - 0x2F */
 	DIP(ModRM | DstMem | Priv | Op3264 | NoMod, cr_read, check_cr_read),
 	DIP(ModRM | DstMem | Priv | Op3264 | NoMod, dr_read, check_dr_read),
@@ -3901,7 +4039,7 @@ static const struct opcode twobyte_table[256] = {
 	N, N, N, N,
 	GP(ModRM | DstReg | SrcMem | Mov | Sse, &pfx_0f_28_0f_29),
 	GP(ModRM | DstMem | SrcReg | Mov | Sse, &pfx_0f_28_0f_29),
-	N, GP(ModRM | DstMem | SrcReg | Sse | Mov | Aligned, &pfx_vmovntpx),
+	N, GP(ModRM | DstMem | SrcReg | Mov | Sse, &pfx_0f_2b),
 	N, N, N, N,
 	/* 0x30 - 0x3F */
 	II(ImplicitOps | Priv, em_wrmsr, wrmsr),
@@ -3942,7 +4080,7 @@ static const struct opcode twobyte_table[256] = {
 	F(DstMem | SrcReg | ModRM | BitOp | Lock | PageTable, em_bts),
 	F(DstMem | SrcReg | Src2ImmByte | ModRM, em_shrd),
 	F(DstMem | SrcReg | Src2CL | ModRM, em_shrd),
-	D(ModRM), F(DstReg | SrcMem | ModRM, em_imul),
+	GD(0, &group15), F(DstReg | SrcMem | ModRM, em_imul),
 	/* 0xB0 - 0xB7 */
 	I2bv(DstMem | SrcReg | ModRM | Lock | PageTable, em_cmpxchg),
 	I(DstReg | SrcMemFAddr | ModRM | Src2SS, em_lseg),
@@ -3965,7 +4103,8 @@ static const struct opcode twobyte_table[256] = {
 	/* 0xD0 - 0xDF */
 	N, N, N, N, N, N, N, N, N, N, N, N, N, N, N, N,
 	/* 0xE0 - 0xEF */
-	N, N, N, N, N, N, N, N, N, N, N, N, N, N, N, N,
+	N, N, N, N, N, N, N, GP(SrcReg | DstMem | ModRM | Mov, &pfx_0f_e7),
+	N, N, N, N, N, N, N, N,
 	/* 0xF0 - 0xFF */
 	N, N, N, N, N, N, N, N, N, N, N, N, N, N, N, N
 };
@@ -4458,10 +4597,10 @@ done_prefixes:
 	/* Decode and fetch the destination operand: register or memory. */
 	rc = decode_operand(ctxt, &ctxt->dst, (ctxt->d >> DstShift) & OpMask);
 
-done:
 	if (ctxt->rip_relative)
 		ctxt->memopp->addr.mem.ea += ctxt->_eip;
 
+done:
 	return (rc != X86EMUL_CONTINUE) ? EMULATION_FAILED : EMULATION_OK;
 }
 
@@ -4711,7 +4850,7 @@ special_insn:
 		break;
 	case 0x70 ... 0x7f: /* jcc (short) */
 		if (test_cc(ctxt->b, ctxt->eflags))
-			jmp_rel(ctxt, ctxt->src.val);
+			rc = jmp_rel(ctxt, ctxt->src.val);
 		break;
 	case 0x8d: /* lea r16/r32, m */
 		ctxt->dst.val = ctxt->src.addr.mem.ea;
@@ -4741,7 +4880,7 @@ special_insn:
 		break;
 	case 0xe9: /* jmp rel */
 	case 0xeb: /* jmp rel short */
-		jmp_rel(ctxt, ctxt->src.val);
+		rc = jmp_rel(ctxt, ctxt->src.val);
 		ctxt->dst.type = OP_NONE; /* Disable writeback. */
 		break;
 	case 0xf4:              /* hlt */
@@ -4829,8 +4968,10 @@ writeback:
 	ctxt->eip = ctxt->_eip;
 
 done:
-	if (rc == X86EMUL_PROPAGATE_FAULT)
+	if (rc == X86EMUL_PROPAGATE_FAULT) {
+		WARN_ON(ctxt->exception.vector > 0x1f);
 		ctxt->have_exception = true;
+	}
 	if (rc == X86EMUL_INTERCEPTED)
 		return EMULATION_INTERCEPTED;
 
@@ -4864,13 +5005,11 @@ twobyte_insn:
 		break;
 	case 0x80 ... 0x8f: /* jnz rel, etc*/
 		if (test_cc(ctxt->b, ctxt->eflags))
-			jmp_rel(ctxt, ctxt->src.val);
+			rc = jmp_rel(ctxt, ctxt->src.val);
 		break;
 	case 0x90 ... 0x9f:     /* setcc r/m8 */
 		ctxt->dst.val = test_cc(ctxt->b, ctxt->eflags);
 		break;
-	case 0xae:              /* clflush */
-		break;
 	case 0xb6 ... 0xb7:	/* movzx */
 		ctxt->dst.bytes = ctxt->op_bytes;
 		ctxt->dst.val = (ctxt->src.bytes == 1) ? (u8) ctxt->src.val