Changeset 62091 in vbox for trunk/src/VBox/VMM/VMMAll

Timestamp:

Jul 7, 2016 12:25:11 AM (9 years ago)

Author:

vboxsync

svn:sync-xref-src-repo-rev:

108564

Message:

IEM: Converted IEMOP_HLP_NO_LOCK_PREFIX to IEMOP_HLP_DONE_DECODING_NO_LOCK_PREFIX and put them after decoded is done. Expect slight code size increase.

Location:

trunk/src/VBox/VMM/VMMAll

Files:

• IEMAll.cpp (modified) (1 diff)
• IEMAllInstructions.cpp.h (modified) (371 diffs)

trunk/src/VBox/VMM/VMMAll/IEMAll.cpp

@@ -10769 +10769 @@
     } while (0)
 
-/** The instruction allows no lock prefixing (in this encoding), throw \#UD if
- * lock prefixed.
- * @deprecated  IEMOP_HLP_DONE_DECODING_NO_LOCK_PREFIX */
-#define IEMOP_HLP_NO_LOCK_PREFIX() \
-    do \
-    { \
-        if (pVCpu->iem.s.fPrefixes & IEM_OP_PRF_LOCK) \
-            return IEMOP_RAISE_INVALID_LOCK_PREFIX(); \
-    } while (0)
-
 /** The instruction is not available in 64-bit mode, throw \#UD if we're in
  * 64-bit mode. */

trunk/src/VBox/VMM/VMMAll/IEMAllInstructions.cpp.h

@@ -38 +38 @@
     if ((bRm & X86_MODRM_MOD_MASK) == (3 << X86_MODRM_MOD_SHIFT))
     {
-        IEMOP_HLP_NO_LOCK_PREFIX();
+        IEMOP_HLP_DONE_DECODING_NO_LOCK_PREFIX();
 
         IEM_MC_BEGIN(3, 0);
@@ -68 +68 @@
 
         IEM_MC_CALC_RM_EFF_ADDR(GCPtrEffDst, bRm, 0);
+        if (!pImpl->pfnLockedU8)
+            IEMOP_HLP_DONE_DECODING_NO_LOCK_PREFIX();
         IEM_MC_MEM_MAP(pu8Dst, fAccess, pVCpu->iem.s.iEffSeg, GCPtrEffDst, 0 /*arg*/);
         IEM_MC_FETCH_GREG_U8(u8Src, ((bRm >> X86_MODRM_REG_SHIFT) & X86_MODRM_REG_SMASK) | pVCpu->iem.s.uRexReg);
@@ -100 +102 @@
     if ((bRm & X86_MODRM_MOD_MASK) == (3 << X86_MODRM_MOD_SHIFT))
     {
-        IEMOP_HLP_NO_LOCK_PREFIX();
+        IEMOP_HLP_DONE_DECODING_NO_LOCK_PREFIX();
 
         switch (pVCpu->iem.s.enmEffOpSize)
@@ -170 +172 @@
 
                 IEM_MC_CALC_RM_EFF_ADDR(GCPtrEffDst, bRm, 0);
+                if (!pImpl->pfnLockedU16)
+                    IEMOP_HLP_DONE_DECODING_NO_LOCK_PREFIX();
                 IEM_MC_MEM_MAP(pu16Dst, fAccess, pVCpu->iem.s.iEffSeg, GCPtrEffDst, 0 /*arg*/);
                 IEM_MC_FETCH_GREG_U16(u16Src, ((bRm >> X86_MODRM_REG_SHIFT) & X86_MODRM_REG_SMASK) | pVCpu->iem.s.uRexReg);
@@ -192 +196 @@
 
                 IEM_MC_CALC_RM_EFF_ADDR(GCPtrEffDst, bRm, 0);
+                if (!pImpl->pfnLockedU32)
+                    IEMOP_HLP_DONE_DECODING_NO_LOCK_PREFIX();
                 IEM_MC_MEM_MAP(pu32Dst, fAccess, pVCpu->iem.s.iEffSeg, GCPtrEffDst, 0 /*arg*/);
                 IEM_MC_FETCH_GREG_U32(u32Src, ((bRm >> X86_MODRM_REG_SHIFT) & X86_MODRM_REG_SMASK) | pVCpu->iem.s.uRexReg);
@@ -214 +220 @@
 
                 IEM_MC_CALC_RM_EFF_ADDR(GCPtrEffDst, bRm, 0);
+                if (!pImpl->pfnLockedU64)
+                    IEMOP_HLP_DONE_DECODING_NO_LOCK_PREFIX();
                 IEM_MC_MEM_MAP(pu64Dst, fAccess, pVCpu->iem.s.iEffSeg, GCPtrEffDst, 0 /*arg*/);
                 IEM_MC_FETCH_GREG_U64(u64Src, ((bRm >> X86_MODRM_REG_SHIFT) & X86_MODRM_REG_SMASK) | pVCpu->iem.s.uRexReg);
@@ -242 +250 @@
 {
     uint8_t bRm; IEM_OPCODE_GET_NEXT_U8(&bRm);
-    IEMOP_HLP_NO_LOCK_PREFIX(); /** @todo should probably not be raised until we've fetched all the opcode bytes? */
 
     /*
@@ -249 +256 @@
     if ((bRm & X86_MODRM_MOD_MASK) == (3 << X86_MODRM_MOD_SHIFT))
     {
+        IEMOP_HLP_DONE_DECODING_NO_LOCK_PREFIX();
         IEM_MC_BEGIN(3, 0);
         IEM_MC_ARG(uint8_t *,  pu8Dst,  0);
@@ -274 +282 @@
 
         IEM_MC_CALC_RM_EFF_ADDR(GCPtrEffDst, bRm, 0);
+        IEMOP_HLP_DONE_DECODING_NO_LOCK_PREFIX();
         IEM_MC_FETCH_MEM_U8(u8Src, pVCpu->iem.s.iEffSeg, GCPtrEffDst);
         IEM_MC_REF_GREG_U8(pu8Dst, ((bRm >> X86_MODRM_REG_SHIFT) & X86_MODRM_REG_SMASK) | pVCpu->iem.s.uRexReg);
@@ -295 +304 @@
 {
     uint8_t bRm; IEM_OPCODE_GET_NEXT_U8(&bRm);
-    IEMOP_HLP_NO_LOCK_PREFIX(); /** @todo should probably not be raised until we've fetched all the opcode bytes? */
 
     /*
@@ -302 +310 @@
     if ((bRm & X86_MODRM_MOD_MASK) == (3 << X86_MODRM_MOD_SHIFT))
     {
+        IEMOP_HLP_DONE_DECODING_NO_LOCK_PREFIX();
         switch (pVCpu->iem.s.enmEffOpSize)
         {
@@ -366 +375 @@
 
                 IEM_MC_CALC_RM_EFF_ADDR(GCPtrEffDst, bRm, 0);
+                IEMOP_HLP_DONE_DECODING_NO_LOCK_PREFIX();
                 IEM_MC_FETCH_MEM_U16(u16Src, pVCpu->iem.s.iEffSeg, GCPtrEffDst);
                 IEM_MC_REF_GREG_U16(pu16Dst, ((bRm >> X86_MODRM_REG_SHIFT) & X86_MODRM_REG_SMASK) | pVCpu->iem.s.uRexReg);
@@ -383 +393 @@
 
                 IEM_MC_CALC_RM_EFF_ADDR(GCPtrEffDst, bRm, 0);
+                IEMOP_HLP_DONE_DECODING_NO_LOCK_PREFIX();
                 IEM_MC_FETCH_MEM_U32(u32Src, pVCpu->iem.s.iEffSeg, GCPtrEffDst);
                 IEM_MC_REF_GREG_U32(pu32Dst, ((bRm >> X86_MODRM_REG_SHIFT) & X86_MODRM_REG_SMASK) | pVCpu->iem.s.uRexReg);
@@ -401 +412 @@
 
                 IEM_MC_CALC_RM_EFF_ADDR(GCPtrEffDst, bRm, 0);
+                IEMOP_HLP_DONE_DECODING_NO_LOCK_PREFIX();
                 IEM_MC_FETCH_MEM_U64(u64Src, pVCpu->iem.s.iEffSeg, GCPtrEffDst);
                 IEM_MC_REF_GREG_U64(pu64Dst, ((bRm >> X86_MODRM_REG_SHIFT) & X86_MODRM_REG_SMASK) | pVCpu->iem.s.uRexReg);
@@ -424 +436 @@
 {
     uint8_t u8Imm; IEM_OPCODE_GET_NEXT_U8(&u8Imm);
-    IEMOP_HLP_NO_LOCK_PREFIX();
+    IEMOP_HLP_DONE_DECODING_NO_LOCK_PREFIX();
 
     IEM_MC_BEGIN(3, 0);
@@ -454 +466 @@
         {
             uint16_t u16Imm; IEM_OPCODE_GET_NEXT_U16(&u16Imm);
-            IEMOP_HLP_NO_LOCK_PREFIX();
+            IEMOP_HLP_DONE_DECODING_NO_LOCK_PREFIX();
 
             IEM_MC_BEGIN(3, 0);
@@ -473 +485 @@
         {
             uint32_t u32Imm; IEM_OPCODE_GET_NEXT_U32(&u32Imm);
-            IEMOP_HLP_NO_LOCK_PREFIX();
+            IEMOP_HLP_DONE_DECODING_NO_LOCK_PREFIX();
 
             IEM_MC_BEGIN(3, 0);
@@ -494 +506 @@
         {
             uint64_t u64Imm; IEM_OPCODE_GET_NEXT_S32_SX_U64(&u64Imm);
-            IEMOP_HLP_NO_LOCK_PREFIX();
+            IEMOP_HLP_DONE_DECODING_NO_LOCK_PREFIX();
 
             IEM_MC_BEGIN(3, 0);
@@ -976 +988 @@
     IEMOP_MNEMONIC("smsw");
     IEMOP_HLP_MIN_286();
-    IEMOP_HLP_NO_LOCK_PREFIX();
     if ((bRm & X86_MODRM_MOD_MASK) == (3 << X86_MODRM_MOD_SHIFT))
     {
+        IEMOP_HLP_DONE_DECODING_NO_LOCK_PREFIX();
         switch (pVCpu->iem.s.enmEffOpSize)
         {
@@ -1024 +1036 @@
         IEM_MC_LOCAL(RTGCPTR,  GCPtrEffDst);
         IEM_MC_CALC_RM_EFF_ADDR(GCPtrEffDst, bRm, 0);
+        IEMOP_HLP_DONE_DECODING_NO_LOCK_PREFIX();
         IEM_MC_FETCH_CR0_U16(u16Tmp);
         if (IEM_GET_TARGET_CPU(pVCpu) > IEMTARGETCPU_386)
@@ -1046 +1059 @@
     IEMOP_MNEMONIC("lmsw");
     IEMOP_HLP_MIN_286();
-    IEMOP_HLP_NO_LOCK_PREFIX();
     if ((bRm & X86_MODRM_MOD_MASK) == (3 << X86_MODRM_MOD_SHIFT))
     {
+        IEMOP_HLP_DONE_DECODING_NO_LOCK_PREFIX();
         IEM_MC_BEGIN(1, 0);
         IEM_MC_ARG(uint16_t, u16Tmp, 0);
@@ -1061 +1074 @@
         IEM_MC_LOCAL(RTGCPTR,  GCPtrEffDst);
         IEM_MC_CALC_RM_EFF_ADDR(GCPtrEffDst, bRm, 0);
+        IEMOP_HLP_DONE_DECODING_NO_LOCK_PREFIX();
         IEM_MC_FETCH_MEM_U16(u16Tmp, pVCpu->iem.s.iEffSeg, GCPtrEffDst);
         IEM_MC_CALL_CIMPL_1(iemCImpl_lmsw, u16Tmp);
@@ -1074 +1088 @@
     IEMOP_MNEMONIC("invlpg");
     IEMOP_HLP_MIN_486();
-    IEMOP_HLP_NO_LOCK_PREFIX();
+    IEMOP_HLP_DONE_DECODING_NO_LOCK_PREFIX();
     IEM_MC_BEGIN(1, 1);
     IEM_MC_ARG(RTGCPTR, GCPtrEffDst, 0);
@@ -1089 +1103 @@
     IEMOP_MNEMONIC("swapgs");
     IEMOP_HLP_ONLY_64BIT();
-    IEMOP_HLP_NO_LOCK_PREFIX();
+    IEMOP_HLP_DONE_DECODING_NO_LOCK_PREFIX();
     return IEM_MC_DEFER_TO_CIMPL_0(iemCImpl_swapgs);
 }
@@ -1304 +1318 @@
 {
     IEMOP_MNEMONIC("syscall"); /** @todo 286 LOADALL   */
-    IEMOP_HLP_NO_LOCK_PREFIX();
+    IEMOP_HLP_DONE_DECODING_NO_LOCK_PREFIX();
     return IEM_MC_DEFER_TO_CIMPL_0(iemCImpl_syscall);
 }
@@ -1313 +1327 @@
 {
     IEMOP_MNEMONIC("clts");
-    IEMOP_HLP_NO_LOCK_PREFIX();
+    IEMOP_HLP_DONE_DECODING_NO_LOCK_PREFIX();
     return IEM_MC_DEFER_TO_CIMPL_0(iemCImpl_clts);
 }
@@ -1322 +1336 @@
 {
     IEMOP_MNEMONIC("sysret");  /** @todo 386 LOADALL   */
-    IEMOP_HLP_NO_LOCK_PREFIX();
+    IEMOP_HLP_DONE_DECODING_NO_LOCK_PREFIX();
     return IEM_MC_DEFER_TO_CIMPL_0(iemCImpl_sysret);
 }
@@ -1337 +1351 @@
     IEMOP_MNEMONIC("wbinvd");
     IEMOP_HLP_MIN_486();
-    IEMOP_HLP_NO_LOCK_PREFIX();
+    IEMOP_HLP_DONE_DECODING_NO_LOCK_PREFIX();
     IEM_MC_BEGIN(0, 0);
     IEM_MC_RAISE_GP0_IF_CPL_NOT_ZERO();
@@ -1370 +1384 @@
     }
 
-    IEMOP_HLP_NO_LOCK_PREFIX();
     switch ((bRm >> X86_MODRM_REG_SHIFT) & X86_MODRM_REG_SMASK)
     {
@@ -1387 +1400 @@
     IEM_MC_LOCAL(RTGCPTR,  GCPtrEffSrc);
     IEM_MC_CALC_RM_EFF_ADDR(GCPtrEffSrc, bRm, 0);
+    IEMOP_HLP_DONE_DECODING_NO_LOCK_PREFIX();
     /* Currently a NOP. */
     IEM_MC_ADVANCE_RIP();
@@ -1688 +1702 @@
     if ((bRm & X86_MODRM_MOD_MASK) != (3 << X86_MODRM_MOD_SHIFT))
     {
-        IEMOP_HLP_NO_LOCK_PREFIX();
         switch ((bRm >> X86_MODRM_REG_SHIFT) & X86_MODRM_REG_SMASK)
         {
@@ -1705 +1718 @@
         IEM_MC_LOCAL(RTGCPTR,  GCPtrEffSrc);
         IEM_MC_CALC_RM_EFF_ADDR(GCPtrEffSrc, bRm, 0);
+        IEMOP_HLP_DONE_DECODING_NO_LOCK_PREFIX();
         /* Currently a NOP. */
         IEM_MC_ADVANCE_RIP();
@@ -1718 +1732 @@
 FNIEMOP_DEF(iemOp_nop_Ev)
 {
-    IEMOP_HLP_NO_LOCK_PREFIX();
     uint8_t bRm; IEM_OPCODE_GET_NEXT_U8(&bRm);
     if ((bRm & X86_MODRM_MOD_MASK) == (3 << X86_MODRM_MOD_SHIFT))
     {
+        IEMOP_HLP_DONE_DECODING_NO_LOCK_PREFIX();
         IEM_MC_BEGIN(0, 0);
         IEM_MC_ADVANCE_RIP();
@@ -1731 +1745 @@
         IEM_MC_LOCAL(RTGCPTR, GCPtrEffSrc);
         IEM_MC_CALC_RM_EFF_ADDR(GCPtrEffSrc, bRm, 0);
+        IEMOP_HLP_DONE_DECODING_NO_LOCK_PREFIX();
         /* Currently a NOP. */
         IEM_MC_ADVANCE_RIP();
@@ -1778 +1793 @@
     IEMOP_HLP_MIN_386();
     uint8_t bRm; IEM_OPCODE_GET_NEXT_U8(&bRm);
-    IEMOP_HLP_NO_LOCK_PREFIX();
+    IEMOP_HLP_DONE_DECODING_NO_LOCK_PREFIX();
     if (pVCpu->iem.s.fPrefixes & IEM_OP_PRF_REX_R)
         return IEMOP_RAISE_INVALID_OPCODE();
@@ -3482 +3497 @@
     {
         int16_t i16Imm; IEM_OPCODE_GET_NEXT_S16(&i16Imm);
-        IEMOP_HLP_NO_LOCK_PREFIX();
+        IEMOP_HLP_DONE_DECODING_NO_LOCK_PREFIX();
 
         IEM_MC_BEGIN(0, 0);
@@ -3495 +3510 @@
     {
         int32_t i32Imm; IEM_OPCODE_GET_NEXT_S32(&i32Imm);
-        IEMOP_HLP_NO_LOCK_PREFIX();
+        IEMOP_HLP_DONE_DECODING_NO_LOCK_PREFIX();
 
         IEM_MC_BEGIN(0, 0);
@@ -3518 +3533 @@
     {
         int16_t i16Imm; IEM_OPCODE_GET_NEXT_S16(&i16Imm);
-        IEMOP_HLP_NO_LOCK_PREFIX();
+        IEMOP_HLP_DONE_DECODING_NO_LOCK_PREFIX();
 
         IEM_MC_BEGIN(0, 0);
@@ -3531 +3546 @@
     {
         int32_t i32Imm; IEM_OPCODE_GET_NEXT_S32(&i32Imm);
-        IEMOP_HLP_NO_LOCK_PREFIX();
+        IEMOP_HLP_DONE_DECODING_NO_LOCK_PREFIX();
 
         IEM_MC_BEGIN(0, 0);
@@ -3554 +3569 @@
     {
         int16_t i16Imm; IEM_OPCODE_GET_NEXT_S16(&i16Imm);
-        IEMOP_HLP_NO_LOCK_PREFIX();
+        IEMOP_HLP_DONE_DECODING_NO_LOCK_PREFIX();
 
         IEM_MC_BEGIN(0, 0);
@@ -3567 +3582 @@
     {
         int32_t i32Imm; IEM_OPCODE_GET_NEXT_S32(&i32Imm);
-        IEMOP_HLP_NO_LOCK_PREFIX();
+        IEMOP_HLP_DONE_DECODING_NO_LOCK_PREFIX();
 
         IEM_MC_BEGIN(0, 0);
@@ -3590 +3605 @@
     {
         int16_t i16Imm; IEM_OPCODE_GET_NEXT_S16(&i16Imm);
-        IEMOP_HLP_NO_LOCK_PREFIX();
+        IEMOP_HLP_DONE_DECODING_NO_LOCK_PREFIX();
 
         IEM_MC_BEGIN(0, 0);
@@ -3603 +3618 @@
     {
         int32_t i32Imm; IEM_OPCODE_GET_NEXT_S32(&i32Imm);
-        IEMOP_HLP_NO_LOCK_PREFIX();
+        IEMOP_HLP_DONE_DECODING_NO_LOCK_PREFIX();
 
         IEM_MC_BEGIN(0, 0);
@@ -3626 +3641 @@
     {
         int16_t i16Imm; IEM_OPCODE_GET_NEXT_S16(&i16Imm);
-        IEMOP_HLP_NO_LOCK_PREFIX();
+        IEMOP_HLP_DONE_DECODING_NO_LOCK_PREFIX();
 
         IEM_MC_BEGIN(0, 0);
@@ -3639 +3654 @@
     {
         int32_t i32Imm; IEM_OPCODE_GET_NEXT_S32(&i32Imm);
-        IEMOP_HLP_NO_LOCK_PREFIX();
+        IEMOP_HLP_DONE_DECODING_NO_LOCK_PREFIX();
 
         IEM_MC_BEGIN(0, 0);
@@ -3662 +3677 @@
     {
         int16_t i16Imm; IEM_OPCODE_GET_NEXT_S16(&i16Imm);
-        IEMOP_HLP_NO_LOCK_PREFIX();
+        IEMOP_HLP_DONE_DECODING_NO_LOCK_PREFIX();
 
         IEM_MC_BEGIN(0, 0);
@@ -3675 +3690 @@
     {
         int32_t i32Imm; IEM_OPCODE_GET_NEXT_S32(&i32Imm);
-        IEMOP_HLP_NO_LOCK_PREFIX();
+        IEMOP_HLP_DONE_DECODING_NO_LOCK_PREFIX();
 
         IEM_MC_BEGIN(0, 0);
@@ -3698 +3713 @@
     {
         int16_t i16Imm; IEM_OPCODE_GET_NEXT_S16(&i16Imm);
-        IEMOP_HLP_NO_LOCK_PREFIX();
+        IEMOP_HLP_DONE_DECODING_NO_LOCK_PREFIX();
 
         IEM_MC_BEGIN(0, 0);
@@ -3711 +3726 @@
     {
         int32_t i32Imm; IEM_OPCODE_GET_NEXT_S32(&i32Imm);
-        IEMOP_HLP_NO_LOCK_PREFIX();
+        IEMOP_HLP_DONE_DECODING_NO_LOCK_PREFIX();
 
         IEM_MC_BEGIN(0, 0);
@@ -3734 +3749 @@
     {
         int16_t i16Imm; IEM_OPCODE_GET_NEXT_S16(&i16Imm);
-        IEMOP_HLP_NO_LOCK_PREFIX();
+        IEMOP_HLP_DONE_DECODING_NO_LOCK_PREFIX();
 
         IEM_MC_BEGIN(0, 0);
@@ -3747 +3762 @@
     {
         int32_t i32Imm; IEM_OPCODE_GET_NEXT_S32(&i32Imm);
-        IEMOP_HLP_NO_LOCK_PREFIX();
+        IEMOP_HLP_DONE_DECODING_NO_LOCK_PREFIX();
 
         IEM_MC_BEGIN(0, 0);
@@ -3770 +3785 @@
     {
         int16_t i16Imm; IEM_OPCODE_GET_NEXT_S16(&i16Imm);
-        IEMOP_HLP_NO_LOCK_PREFIX();
+        IEMOP_HLP_DONE_DECODING_NO_LOCK_PREFIX();
 
         IEM_MC_BEGIN(0, 0);
@@ -3783 +3798 @@
     {
         int32_t i32Imm; IEM_OPCODE_GET_NEXT_S32(&i32Imm);
-        IEMOP_HLP_NO_LOCK_PREFIX();
+        IEMOP_HLP_DONE_DECODING_NO_LOCK_PREFIX();
 
         IEM_MC_BEGIN(0, 0);
@@ -3806 +3821 @@
     {
         int16_t i16Imm; IEM_OPCODE_GET_NEXT_S16(&i16Imm);
-        IEMOP_HLP_NO_LOCK_PREFIX();
+        IEMOP_HLP_DONE_DECODING_NO_LOCK_PREFIX();
 
         IEM_MC_BEGIN(0, 0);
@@ -3819 +3834 @@
     {
         int32_t i32Imm; IEM_OPCODE_GET_NEXT_S32(&i32Imm);
-        IEMOP_HLP_NO_LOCK_PREFIX();
+        IEMOP_HLP_DONE_DECODING_NO_LOCK_PREFIX();
 
         IEM_MC_BEGIN(0, 0);
@@ -3842 +3857 @@
     {
         int16_t i16Imm; IEM_OPCODE_GET_NEXT_S16(&i16Imm);
-        IEMOP_HLP_NO_LOCK_PREFIX();
+        IEMOP_HLP_DONE_DECODING_NO_LOCK_PREFIX();
 
         IEM_MC_BEGIN(0, 0);
@@ -3855 +3870 @@
     {
         int32_t i32Imm; IEM_OPCODE_GET_NEXT_S32(&i32Imm);
-        IEMOP_HLP_NO_LOCK_PREFIX();
+        IEMOP_HLP_DONE_DECODING_NO_LOCK_PREFIX();
 
         IEM_MC_BEGIN(0, 0);
@@ -3878 +3893 @@
     {
         int16_t i16Imm; IEM_OPCODE_GET_NEXT_S16(&i16Imm);
-        IEMOP_HLP_NO_LOCK_PREFIX();
+        IEMOP_HLP_DONE_DECODING_NO_LOCK_PREFIX();
 
         IEM_MC_BEGIN(0, 0);
@@ -3891 +3906 @@
     {
         int32_t i32Imm; IEM_OPCODE_GET_NEXT_S32(&i32Imm);
-        IEMOP_HLP_NO_LOCK_PREFIX();
+        IEMOP_HLP_DONE_DECODING_NO_LOCK_PREFIX();
 
         IEM_MC_BEGIN(0, 0);
@@ -3914 +3929 @@
     {
         int16_t i16Imm; IEM_OPCODE_GET_NEXT_S16(&i16Imm);
-        IEMOP_HLP_NO_LOCK_PREFIX();
+        IEMOP_HLP_DONE_DECODING_NO_LOCK_PREFIX();
 
         IEM_MC_BEGIN(0, 0);
@@ -3927 +3942 @@
     {
         int32_t i32Imm; IEM_OPCODE_GET_NEXT_S32(&i32Imm);
-        IEMOP_HLP_NO_LOCK_PREFIX();
+        IEMOP_HLP_DONE_DECODING_NO_LOCK_PREFIX();
 
         IEM_MC_BEGIN(0, 0);
@@ -3950 +3965 @@
     {
         int16_t i16Imm; IEM_OPCODE_GET_NEXT_S16(&i16Imm);
-        IEMOP_HLP_NO_LOCK_PREFIX();
+        IEMOP_HLP_DONE_DECODING_NO_LOCK_PREFIX();
 
         IEM_MC_BEGIN(0, 0);
@@ -3963 +3978 @@
     {
         int32_t i32Imm; IEM_OPCODE_GET_NEXT_S32(&i32Imm);
-        IEMOP_HLP_NO_LOCK_PREFIX();
+        IEMOP_HLP_DONE_DECODING_NO_LOCK_PREFIX();
 
         IEM_MC_BEGIN(0, 0);
@@ -3986 +4001 @@
     {
         int16_t i16Imm; IEM_OPCODE_GET_NEXT_S16(&i16Imm);
-        IEMOP_HLP_NO_LOCK_PREFIX();
+        IEMOP_HLP_DONE_DECODING_NO_LOCK_PREFIX();
 
         IEM_MC_BEGIN(0, 0);
@@ -3999 +4014 @@
     {
         int32_t i32Imm; IEM_OPCODE_GET_NEXT_S32(&i32Imm);
-        IEMOP_HLP_NO_LOCK_PREFIX();
+        IEMOP_HLP_DONE_DECODING_NO_LOCK_PREFIX();
 
         IEM_MC_BEGIN(0, 0);
@@ -4022 +4037 @@
     {
         int16_t i16Imm; IEM_OPCODE_GET_NEXT_S16(&i16Imm);
-        IEMOP_HLP_NO_LOCK_PREFIX();
+        IEMOP_HLP_DONE_DECODING_NO_LOCK_PREFIX();
 
         IEM_MC_BEGIN(0, 0);
@@ -4035 +4050 @@
     {
         int32_t i32Imm; IEM_OPCODE_GET_NEXT_S32(&i32Imm);
-        IEMOP_HLP_NO_LOCK_PREFIX();
+        IEMOP_HLP_DONE_DECODING_NO_LOCK_PREFIX();
 
         IEM_MC_BEGIN(0, 0);
@@ -4055 +4070 @@
     IEMOP_HLP_MIN_386();
     uint8_t bRm; IEM_OPCODE_GET_NEXT_U8(&bRm);
-    IEMOP_HLP_NO_LOCK_PREFIX(); /** @todo too early? */
 
     /** @todo Encoding test: Check if the 'reg' field is ignored or decoded in
@@ -4063 +4077 @@
     {
         /* register target */
+        IEMOP_HLP_DONE_DECODING_NO_LOCK_PREFIX();
         IEM_MC_BEGIN(0, 0);
         IEM_MC_IF_EFL_BIT_SET(X86_EFL_OF) {
@@ -4078 +4093 @@
         IEM_MC_LOCAL(RTGCPTR, GCPtrEffDst);
         IEM_MC_CALC_RM_EFF_ADDR(GCPtrEffDst, bRm, 0);
+        IEMOP_HLP_DONE_DECODING_NO_LOCK_PREFIX();
         IEM_MC_IF_EFL_BIT_SET(X86_EFL_OF) {
             IEM_MC_STORE_MEM_U8_CONST(pVCpu->iem.s.iEffSeg, GCPtrEffDst, 1);
@@ -4096 +4112 @@
     IEMOP_HLP_MIN_386();
     uint8_t bRm; IEM_OPCODE_GET_NEXT_U8(&bRm);
-    IEMOP_HLP_NO_LOCK_PREFIX(); /** @todo too early? */
 
     /** @todo Encoding test: Check if the 'reg' field is ignored or decoded in
@@ -4104 +4119 @@
     {
         /* register target */
+        IEMOP_HLP_DONE_DECODING_NO_LOCK_PREFIX();
         IEM_MC_BEGIN(0, 0);
         IEM_MC_IF_EFL_BIT_SET(X86_EFL_OF) {
@@ -4119 +4135 @@
         IEM_MC_LOCAL(RTGCPTR, GCPtrEffDst);
         IEM_MC_CALC_RM_EFF_ADDR(GCPtrEffDst, bRm, 0);
+        IEMOP_HLP_DONE_DECODING_NO_LOCK_PREFIX();
         IEM_MC_IF_EFL_BIT_SET(X86_EFL_OF) {
             IEM_MC_STORE_MEM_U8_CONST(pVCpu->iem.s.iEffSeg, GCPtrEffDst, 0);
@@ -4137 +4154 @@
     IEMOP_HLP_MIN_386();
     uint8_t bRm; IEM_OPCODE_GET_NEXT_U8(&bRm);
-    IEMOP_HLP_NO_LOCK_PREFIX(); /** @todo too early? */
 
     /** @todo Encoding test: Check if the 'reg' field is ignored or decoded in
@@ -4145 +4161 @@
     {
         /* register target */
+        IEMOP_HLP_DONE_DECODING_NO_LOCK_PREFIX();
         IEM_MC_BEGIN(0, 0);
         IEM_MC_IF_EFL_BIT_SET(X86_EFL_CF) {
@@ -4160 +4177 @@
         IEM_MC_LOCAL(RTGCPTR, GCPtrEffDst);
         IEM_MC_CALC_RM_EFF_ADDR(GCPtrEffDst, bRm, 0);
+        IEMOP_HLP_DONE_DECODING_NO_LOCK_PREFIX();
         IEM_MC_IF_EFL_BIT_SET(X86_EFL_CF) {
             IEM_MC_STORE_MEM_U8_CONST(pVCpu->iem.s.iEffSeg, GCPtrEffDst, 1);
@@ -4178 +4196 @@
     IEMOP_HLP_MIN_386();
     uint8_t bRm; IEM_OPCODE_GET_NEXT_U8(&bRm);
-    IEMOP_HLP_NO_LOCK_PREFIX(); /** @todo too early? */
 
     /** @todo Encoding test: Check if the 'reg' field is ignored or decoded in
@@ -4186 +4203 @@
     {
         /* register target */
+        IEMOP_HLP_DONE_DECODING_NO_LOCK_PREFIX();
         IEM_MC_BEGIN(0, 0);
         IEM_MC_IF_EFL_BIT_SET(X86_EFL_CF) {
@@ -4201 +4219 @@
         IEM_MC_LOCAL(RTGCPTR, GCPtrEffDst);
         IEM_MC_CALC_RM_EFF_ADDR(GCPtrEffDst, bRm, 0);
+        IEMOP_HLP_DONE_DECODING_NO_LOCK_PREFIX();
         IEM_MC_IF_EFL_BIT_SET(X86_EFL_CF) {
             IEM_MC_STORE_MEM_U8_CONST(pVCpu->iem.s.iEffSeg, GCPtrEffDst, 0);
@@ -4219 +4238 @@
     IEMOP_HLP_MIN_386();
     uint8_t bRm; IEM_OPCODE_GET_NEXT_U8(&bRm);
-    IEMOP_HLP_NO_LOCK_PREFIX(); /** @todo too early? */
 
     /** @todo Encoding test: Check if the 'reg' field is ignored or decoded in
@@ -4227 +4245 @@
     {
         /* register target */
+        IEMOP_HLP_DONE_DECODING_NO_LOCK_PREFIX();
         IEM_MC_BEGIN(0, 0);
         IEM_MC_IF_EFL_BIT_SET(X86_EFL_ZF) {
@@ -4242 +4261 @@
         IEM_MC_LOCAL(RTGCPTR, GCPtrEffDst);
         IEM_MC_CALC_RM_EFF_ADDR(GCPtrEffDst, bRm, 0);
+        IEMOP_HLP_DONE_DECODING_NO_LOCK_PREFIX();
         IEM_MC_IF_EFL_BIT_SET(X86_EFL_ZF) {
             IEM_MC_STORE_MEM_U8_CONST(pVCpu->iem.s.iEffSeg, GCPtrEffDst, 1);
@@ -4260 +4280 @@
     IEMOP_HLP_MIN_386();
     uint8_t bRm; IEM_OPCODE_GET_NEXT_U8(&bRm);
-    IEMOP_HLP_NO_LOCK_PREFIX(); /** @todo too early? */
 
     /** @todo Encoding test: Check if the 'reg' field is ignored or decoded in
@@ -4268 +4287 @@
     {
         /* register target */
+        IEMOP_HLP_DONE_DECODING_NO_LOCK_PREFIX();
         IEM_MC_BEGIN(0, 0);
         IEM_MC_IF_EFL_BIT_SET(X86_EFL_ZF) {
@@ -4283 +4303 @@
         IEM_MC_LOCAL(RTGCPTR, GCPtrEffDst);
         IEM_MC_CALC_RM_EFF_ADDR(GCPtrEffDst, bRm, 0);
+        IEMOP_HLP_DONE_DECODING_NO_LOCK_PREFIX();
         IEM_MC_IF_EFL_BIT_SET(X86_EFL_ZF) {
             IEM_MC_STORE_MEM_U8_CONST(pVCpu->iem.s.iEffSeg, GCPtrEffDst, 0);
@@ -4301 +4322 @@
     IEMOP_HLP_MIN_386();
     uint8_t bRm; IEM_OPCODE_GET_NEXT_U8(&bRm);
-    IEMOP_HLP_NO_LOCK_PREFIX(); /** @todo too early? */
 
     /** @todo Encoding test: Check if the 'reg' field is ignored or decoded in
@@ -4309 +4329 @@
     {
         /* register target */
+        IEMOP_HLP_DONE_DECODING_NO_LOCK_PREFIX();
         IEM_MC_BEGIN(0, 0);
         IEM_MC_IF_EFL_ANY_BITS_SET(X86_EFL_CF | X86_EFL_ZF) {
@@ -4324 +4345 @@
         IEM_MC_LOCAL(RTGCPTR, GCPtrEffDst);
         IEM_MC_CALC_RM_EFF_ADDR(GCPtrEffDst, bRm, 0);
+        IEMOP_HLP_DONE_DECODING_NO_LOCK_PREFIX();
         IEM_MC_IF_EFL_ANY_BITS_SET(X86_EFL_CF | X86_EFL_ZF) {
             IEM_MC_STORE_MEM_U8_CONST(pVCpu->iem.s.iEffSeg, GCPtrEffDst, 1);
@@ -4342 +4364 @@
     IEMOP_HLP_MIN_386();
     uint8_t bRm; IEM_OPCODE_GET_NEXT_U8(&bRm);
-    IEMOP_HLP_NO_LOCK_PREFIX(); /** @todo too early? */
 
     /** @todo Encoding test: Check if the 'reg' field is ignored or decoded in
@@ -4350 +4371 @@
     {
         /* register target */
+        IEMOP_HLP_DONE_DECODING_NO_LOCK_PREFIX();
         IEM_MC_BEGIN(0, 0);
         IEM_MC_IF_EFL_ANY_BITS_SET(X86_EFL_CF | X86_EFL_ZF) {
@@ -4365 +4387 @@
         IEM_MC_LOCAL(RTGCPTR, GCPtrEffDst);
         IEM_MC_CALC_RM_EFF_ADDR(GCPtrEffDst, bRm, 0);
+        IEMOP_HLP_DONE_DECODING_NO_LOCK_PREFIX();
         IEM_MC_IF_EFL_ANY_BITS_SET(X86_EFL_CF | X86_EFL_ZF) {
             IEM_MC_STORE_MEM_U8_CONST(pVCpu->iem.s.iEffSeg, GCPtrEffDst, 0);
@@ -4383 +4406 @@
     IEMOP_HLP_MIN_386();
     uint8_t bRm; IEM_OPCODE_GET_NEXT_U8(&bRm);
-    IEMOP_HLP_NO_LOCK_PREFIX(); /** @todo too early? */
 
     /** @todo Encoding test: Check if the 'reg' field is ignored or decoded in
@@ -4391 +4413 @@
     {
         /* register target */
+        IEMOP_HLP_DONE_DECODING_NO_LOCK_PREFIX();
         IEM_MC_BEGIN(0, 0);
         IEM_MC_IF_EFL_BIT_SET(X86_EFL_SF) {
@@ -4406 +4429 @@
         IEM_MC_LOCAL(RTGCPTR, GCPtrEffDst);
         IEM_MC_CALC_RM_EFF_ADDR(GCPtrEffDst, bRm, 0);
+        IEMOP_HLP_DONE_DECODING_NO_LOCK_PREFIX();
         IEM_MC_IF_EFL_BIT_SET(X86_EFL_SF) {
             IEM_MC_STORE_MEM_U8_CONST(pVCpu->iem.s.iEffSeg, GCPtrEffDst, 1);
@@ -4424 +4448 @@
     IEMOP_HLP_MIN_386();
     uint8_t bRm; IEM_OPCODE_GET_NEXT_U8(&bRm);
-    IEMOP_HLP_NO_LOCK_PREFIX(); /** @todo too early? */
 
     /** @todo Encoding test: Check if the 'reg' field is ignored or decoded in
@@ -4432 +4455 @@
     {
         /* register target */
+        IEMOP_HLP_DONE_DECODING_NO_LOCK_PREFIX();
         IEM_MC_BEGIN(0, 0);
         IEM_MC_IF_EFL_BIT_SET(X86_EFL_SF) {
@@ -4447 +4471 @@
         IEM_MC_LOCAL(RTGCPTR, GCPtrEffDst);
         IEM_MC_CALC_RM_EFF_ADDR(GCPtrEffDst, bRm, 0);
+        IEMOP_HLP_DONE_DECODING_NO_LOCK_PREFIX();
         IEM_MC_IF_EFL_BIT_SET(X86_EFL_SF) {
             IEM_MC_STORE_MEM_U8_CONST(pVCpu->iem.s.iEffSeg, GCPtrEffDst, 0);
@@ -4465 +4490 @@
     IEMOP_HLP_MIN_386();
     uint8_t bRm; IEM_OPCODE_GET_NEXT_U8(&bRm);
-    IEMOP_HLP_NO_LOCK_PREFIX(); /** @todo too early? */
 
     /** @todo Encoding test: Check if the 'reg' field is ignored or decoded in
@@ -4473 +4497 @@
     {
         /* register target */
+        IEMOP_HLP_DONE_DECODING_NO_LOCK_PREFIX();
         IEM_MC_BEGIN(0, 0);
         IEM_MC_IF_EFL_BIT_SET(X86_EFL_PF) {
@@ -4488 +4513 @@
         IEM_MC_LOCAL(RTGCPTR, GCPtrEffDst);
         IEM_MC_CALC_RM_EFF_ADDR(GCPtrEffDst, bRm, 0);
+        IEMOP_HLP_DONE_DECODING_NO_LOCK_PREFIX();
         IEM_MC_IF_EFL_BIT_SET(X86_EFL_PF) {
             IEM_MC_STORE_MEM_U8_CONST(pVCpu->iem.s.iEffSeg, GCPtrEffDst, 1);
@@ -4506 +4532 @@
     IEMOP_HLP_MIN_386();
     uint8_t bRm; IEM_OPCODE_GET_NEXT_U8(&bRm);
-    IEMOP_HLP_NO_LOCK_PREFIX(); /** @todo too early? */
 
     /** @todo Encoding test: Check if the 'reg' field is ignored or decoded in
@@ -4514 +4539 @@
     {
         /* register target */
+        IEMOP_HLP_DONE_DECODING_NO_LOCK_PREFIX();
         IEM_MC_BEGIN(0, 0);
         IEM_MC_IF_EFL_BIT_SET(X86_EFL_PF) {
@@ -4529 +4555 @@
         IEM_MC_LOCAL(RTGCPTR, GCPtrEffDst);
         IEM_MC_CALC_RM_EFF_ADDR(GCPtrEffDst, bRm, 0);
+        IEMOP_HLP_DONE_DECODING_NO_LOCK_PREFIX();
         IEM_MC_IF_EFL_BIT_SET(X86_EFL_PF) {
             IEM_MC_STORE_MEM_U8_CONST(pVCpu->iem.s.iEffSeg, GCPtrEffDst, 0);
@@ -4547 +4574 @@
     IEMOP_HLP_MIN_386();
     uint8_t bRm; IEM_OPCODE_GET_NEXT_U8(&bRm);
-    IEMOP_HLP_NO_LOCK_PREFIX(); /** @todo too early? */
 
     /** @todo Encoding test: Check if the 'reg' field is ignored or decoded in
@@ -4555 +4581 @@
     {
         /* register target */
+        IEMOP_HLP_DONE_DECODING_NO_LOCK_PREFIX();
         IEM_MC_BEGIN(0, 0);
         IEM_MC_IF_EFL_BITS_NE(X86_EFL_SF, X86_EFL_OF) {
@@ -4570 +4597 @@
         IEM_MC_LOCAL(RTGCPTR, GCPtrEffDst);
         IEM_MC_CALC_RM_EFF_ADDR(GCPtrEffDst, bRm, 0);
+        IEMOP_HLP_DONE_DECODING_NO_LOCK_PREFIX();
         IEM_MC_IF_EFL_BITS_NE(X86_EFL_SF, X86_EFL_OF) {
             IEM_MC_STORE_MEM_U8_CONST(pVCpu->iem.s.iEffSeg, GCPtrEffDst, 1);
@@ -4588 +4616 @@
     IEMOP_HLP_MIN_386();
     uint8_t bRm; IEM_OPCODE_GET_NEXT_U8(&bRm);
-    IEMOP_HLP_NO_LOCK_PREFIX(); /** @todo too early? */
 
     /** @todo Encoding test: Check if the 'reg' field is ignored or decoded in
@@ -4596 +4623 @@
     {
         /* register target */
+        IEMOP_HLP_DONE_DECODING_NO_LOCK_PREFIX();
         IEM_MC_BEGIN(0, 0);
         IEM_MC_IF_EFL_BITS_NE(X86_EFL_SF, X86_EFL_OF) {
@@ -4611 +4639 @@
         IEM_MC_LOCAL(RTGCPTR, GCPtrEffDst);
         IEM_MC_CALC_RM_EFF_ADDR(GCPtrEffDst, bRm, 0);
+        IEMOP_HLP_DONE_DECODING_NO_LOCK_PREFIX();
         IEM_MC_IF_EFL_BITS_NE(X86_EFL_SF, X86_EFL_OF) {
             IEM_MC_STORE_MEM_U8_CONST(pVCpu->iem.s.iEffSeg, GCPtrEffDst, 0);
@@ -4629 +4658 @@
     IEMOP_HLP_MIN_386();
     uint8_t bRm; IEM_OPCODE_GET_NEXT_U8(&bRm);
-    IEMOP_HLP_NO_LOCK_PREFIX(); /** @todo too early? */
 
     /** @todo Encoding test: Check if the 'reg' field is ignored or decoded in
@@ -4637 +4665 @@
     {
         /* register target */
+        IEMOP_HLP_DONE_DECODING_NO_LOCK_PREFIX();
         IEM_MC_BEGIN(0, 0);
         IEM_MC_IF_EFL_BIT_SET_OR_BITS_NE(X86_EFL_ZF, X86_EFL_SF, X86_EFL_OF) {
@@ -4652 +4681 @@
         IEM_MC_LOCAL(RTGCPTR, GCPtrEffDst);
         IEM_MC_CALC_RM_EFF_ADDR(GCPtrEffDst, bRm, 0);
+        IEMOP_HLP_DONE_DECODING_NO_LOCK_PREFIX();
         IEM_MC_IF_EFL_BIT_SET_OR_BITS_NE(X86_EFL_ZF, X86_EFL_SF, X86_EFL_OF) {
             IEM_MC_STORE_MEM_U8_CONST(pVCpu->iem.s.iEffSeg, GCPtrEffDst, 1);
@@ -4670 +4700 @@
     IEMOP_HLP_MIN_386();
     uint8_t bRm; IEM_OPCODE_GET_NEXT_U8(&bRm);
-    IEMOP_HLP_NO_LOCK_PREFIX(); /** @todo too early? */
 
     /** @todo Encoding test: Check if the 'reg' field is ignored or decoded in
@@ -4678 +4707 @@
     {
         /* register target */
+        IEMOP_HLP_DONE_DECODING_NO_LOCK_PREFIX();
         IEM_MC_BEGIN(0, 0);
         IEM_MC_IF_EFL_BIT_SET_OR_BITS_NE(X86_EFL_ZF, X86_EFL_SF, X86_EFL_OF) {
@@ -4693 +4723 @@
         IEM_MC_LOCAL(RTGCPTR, GCPtrEffDst);
         IEM_MC_CALC_RM_EFF_ADDR(GCPtrEffDst, bRm, 0);
+        IEMOP_HLP_DONE_DECODING_NO_LOCK_PREFIX();
         IEM_MC_IF_EFL_BIT_SET_OR_BITS_NE(X86_EFL_ZF, X86_EFL_SF, X86_EFL_OF) {
             IEM_MC_STORE_MEM_U8_CONST(pVCpu->iem.s.iEffSeg, GCPtrEffDst, 0);
@@ -4710 +4741 @@
 FNIEMOP_DEF_1(iemOpCommonPushSReg, uint8_t, iReg)
 {
-    IEMOP_HLP_NO_LOCK_PREFIX();
+    IEMOP_HLP_DONE_DECODING_NO_LOCK_PREFIX();
     if (iReg < X86_SREG_FS)
         IEMOP_HLP_NO_64BIT();
@@ -4754 +4785 @@
     IEMOP_MNEMONIC("push fs");
     IEMOP_HLP_MIN_386();
-    IEMOP_HLP_NO_LOCK_PREFIX();
+    IEMOP_HLP_DONE_DECODING_NO_LOCK_PREFIX();
     return FNIEMOP_CALL_1(iemOpCommonPushSReg, X86_SREG_FS);
 }
@@ -4764 +4795 @@
     IEMOP_MNEMONIC("pop fs");
     IEMOP_HLP_MIN_386();
-    IEMOP_HLP_NO_LOCK_PREFIX();
+    IEMOP_HLP_DONE_DECODING_NO_LOCK_PREFIX();
     return IEM_MC_DEFER_TO_CIMPL_2(iemCImpl_pop_Sreg, X86_SREG_FS, pVCpu->iem.s.enmEffOpSize);
 }
@@ -4774 +4805 @@
     IEMOP_MNEMONIC("cpuid");
     IEMOP_HLP_MIN_486(); /* not all 486es. */
-    IEMOP_HLP_NO_LOCK_PREFIX();
+    IEMOP_HLP_DONE_DECODING_NO_LOCK_PREFIX();
     return IEM_MC_DEFER_TO_CIMPL_0(iemCImpl_cpuid);
 }
@@ -4791 +4822 @@
     {
         /* register destination. */
-        IEMOP_HLP_NO_LOCK_PREFIX();
+        IEMOP_HLP_DONE_DECODING_NO_LOCK_PREFIX();
         switch (pVCpu->iem.s.enmEffOpSize)
         {
@@ -4854 +4885 @@
             fAccess = IEM_ACCESS_DATA_RW;
         else /* BT */
-        {
-            IEMOP_HLP_NO_LOCK_PREFIX();
             fAccess = IEM_ACCESS_DATA_R;
-        }
 
         NOREF(fAccess);
@@ -4873 +4901 @@
 
                 IEM_MC_CALC_RM_EFF_ADDR(GCPtrEffDst, bRm, 0);
+                if (pImpl->pfnLockedU16)
+                    IEMOP_HLP_DONE_DECODING();
+                else
+                    IEMOP_HLP_DONE_DECODING_NO_LOCK_PREFIX();
                 IEM_MC_FETCH_GREG_U16(u16Src, ((bRm >> X86_MODRM_REG_SHIFT) & X86_MODRM_REG_SMASK) | pVCpu->iem.s.uRexReg);
                 IEM_MC_ASSIGN(i16AddrAdj, u16Src);
@@ -4902 +4934 @@
 
                 IEM_MC_CALC_RM_EFF_ADDR(GCPtrEffDst, bRm, 0);
+                if (pImpl->pfnLockedU16)
+                    IEMOP_HLP_DONE_DECODING();
+                else
+                    IEMOP_HLP_DONE_DECODING_NO_LOCK_PREFIX();
                 IEM_MC_FETCH_GREG_U32(u32Src, ((bRm >> X86_MODRM_REG_SHIFT) & X86_MODRM_REG_SMASK) | pVCpu->iem.s.uRexReg);
                 IEM_MC_ASSIGN(i32AddrAdj, u32Src);
@@ -4931 +4967 @@
 
                 IEM_MC_CALC_RM_EFF_ADDR(GCPtrEffDst, bRm, 0);
+                if (pImpl->pfnLockedU16)
+                    IEMOP_HLP_DONE_DECODING();
+                else
+                    IEMOP_HLP_DONE_DECODING_NO_LOCK_PREFIX();
                 IEM_MC_FETCH_GREG_U64(u64Src, ((bRm >> X86_MODRM_REG_SHIFT) & X86_MODRM_REG_SMASK) | pVCpu->iem.s.uRexReg);
                 IEM_MC_ASSIGN(i64AddrAdj, u64Src);
@@ -4972 +5012 @@
 {
     uint8_t bRm; IEM_OPCODE_GET_NEXT_U8(&bRm);
-    IEMOP_HLP_NO_LOCK_PREFIX();
     IEMOP_VERIFICATION_UNDEFINED_EFLAGS(X86_EFL_AF | X86_EFL_OF);
 
@@ -4978 +5017 @@
     {
         uint8_t cShift; IEM_OPCODE_GET_NEXT_U8(&cShift);
-        IEMOP_HLP_NO_LOCK_PREFIX();
+        IEMOP_HLP_DONE_DECODING_NO_LOCK_PREFIX();
 
         switch (pVCpu->iem.s.enmEffOpSize)
@@ -5036 +5075 @@
     else
     {
-        IEMOP_HLP_NO_LOCK_PREFIX(); /** @todo too early? */
-
         switch (pVCpu->iem.s.enmEffOpSize)
         {
@@ -5051 +5088 @@
                 uint8_t cShift; IEM_OPCODE_GET_NEXT_U8(&cShift);
                 IEM_MC_ASSIGN(cShiftArg, cShift);
+                IEMOP_HLP_DONE_DECODING_NO_LOCK_PREFIX();
                 IEM_MC_FETCH_GREG_U16(u16Src, ((bRm >> X86_MODRM_REG_SHIFT) & X86_MODRM_REG_SMASK) | pVCpu->iem.s.uRexReg);
                 IEM_MC_FETCH_EFLAGS(EFlags);
@@ -5073 +5111 @@
                 uint8_t cShift; IEM_OPCODE_GET_NEXT_U8(&cShift);
                 IEM_MC_ASSIGN(cShiftArg, cShift);
+                IEMOP_HLP_DONE_DECODING_NO_LOCK_PREFIX();
                 IEM_MC_FETCH_GREG_U32(u32Src, ((bRm >> X86_MODRM_REG_SHIFT) & X86_MODRM_REG_SMASK) | pVCpu->iem.s.uRexReg);
                 IEM_MC_FETCH_EFLAGS(EFlags);
@@ -5095 +5134 @@
                 uint8_t cShift; IEM_OPCODE_GET_NEXT_U8(&cShift);
                 IEM_MC_ASSIGN(cShiftArg, cShift);
+                IEMOP_HLP_DONE_DECODING_NO_LOCK_PREFIX();
                 IEM_MC_FETCH_GREG_U64(u64Src, ((bRm >> X86_MODRM_REG_SHIFT) & X86_MODRM_REG_SMASK) | pVCpu->iem.s.uRexReg);
                 IEM_MC_FETCH_EFLAGS(EFlags);
@@ -5118 +5158 @@
 {
     uint8_t bRm; IEM_OPCODE_GET_NEXT_U8(&bRm);
-    IEMOP_HLP_NO_LOCK_PREFIX();
     IEMOP_VERIFICATION_UNDEFINED_EFLAGS(X86_EFL_AF | X86_EFL_OF);
 
     if ((bRm & X86_MODRM_MOD_MASK) == (3 << X86_MODRM_MOD_SHIFT))
     {
-        IEMOP_HLP_NO_LOCK_PREFIX();
+        IEMOP_HLP_DONE_DECODING_NO_LOCK_PREFIX();
 
         switch (pVCpu->iem.s.enmEffOpSize)
@@ -5184 +5223 @@
     else
     {
-        IEMOP_HLP_NO_LOCK_PREFIX(); /** @todo too early? */
-
         switch (pVCpu->iem.s.enmEffOpSize)
         {
@@ -5197 +5234 @@
 
                 IEM_MC_CALC_RM_EFF_ADDR(GCPtrEffDst, bRm, 0);
+                IEMOP_HLP_DONE_DECODING_NO_LOCK_PREFIX();
                 IEM_MC_FETCH_GREG_U16(u16Src, ((bRm >> X86_MODRM_REG_SHIFT) & X86_MODRM_REG_SMASK) | pVCpu->iem.s.uRexReg);
                 IEM_MC_FETCH_GREG_U8(cShiftArg, X86_GREG_xCX);
@@ -5218 +5256 @@
 
                 IEM_MC_CALC_RM_EFF_ADDR(GCPtrEffDst, bRm, 0);
+                IEMOP_HLP_DONE_DECODING_NO_LOCK_PREFIX();
                 IEM_MC_FETCH_GREG_U32(u32Src, ((bRm >> X86_MODRM_REG_SHIFT) & X86_MODRM_REG_SMASK) | pVCpu->iem.s.uRexReg);
                 IEM_MC_FETCH_GREG_U8(cShiftArg, X86_GREG_xCX);
@@ -5239 +5278 @@
 
                 IEM_MC_CALC_RM_EFF_ADDR(GCPtrEffDst, bRm, 0);
+                IEMOP_HLP_DONE_DECODING_NO_LOCK_PREFIX();
                 IEM_MC_FETCH_GREG_U64(u64Src, ((bRm >> X86_MODRM_REG_SHIFT) & X86_MODRM_REG_SMASK) | pVCpu->iem.s.uRexReg);
                 IEM_MC_FETCH_GREG_U8(cShiftArg, X86_GREG_xCX);
@@ -5281 +5321 @@
     IEMOP_MNEMONIC("push gs");
     IEMOP_HLP_MIN_386();
-    IEMOP_HLP_NO_LOCK_PREFIX();
+    IEMOP_HLP_DONE_DECODING_NO_LOCK_PREFIX();
     return FNIEMOP_CALL_1(iemOpCommonPushSReg, X86_SREG_GS);
 }
@@ -5291 +5331 @@
     IEMOP_MNEMONIC("pop gs");
     IEMOP_HLP_MIN_386();
-    IEMOP_HLP_NO_LOCK_PREFIX();
+    IEMOP_HLP_DONE_DECODING_NO_LOCK_PREFIX();
     return IEM_MC_DEFER_TO_CIMPL_2(iemCImpl_pop_Sreg, X86_SREG_GS, pVCpu->iem.s.enmEffOpSize);
 }
@@ -5391 +5431 @@
 {
     IEMOP_MNEMONIC("lfence");
-    IEMOP_HLP_NO_LOCK_PREFIX();
+    IEMOP_HLP_DONE_DECODING_NO_LOCK_PREFIX();
     if (!IEM_GET_GUEST_CPU_FEATURES(pVCpu)->fSse2)
         return IEMOP_RAISE_INVALID_OPCODE();
@@ -5410 +5450 @@
 {
     IEMOP_MNEMONIC("mfence");
-    IEMOP_HLP_NO_LOCK_PREFIX();
+    IEMOP_HLP_DONE_DECODING_NO_LOCK_PREFIX();
     if (!IEM_GET_GUEST_CPU_FEATURES(pVCpu)->fSse2)
         return IEMOP_RAISE_INVALID_OPCODE();
@@ -5429 +5469 @@
 {
     IEMOP_MNEMONIC("sfence");
-    IEMOP_HLP_NO_LOCK_PREFIX();
+    IEMOP_HLP_DONE_DECODING_NO_LOCK_PREFIX();
     if (!IEM_GET_GUEST_CPU_FEATURES(pVCpu)->fSse2)
         return IEMOP_RAISE_INVALID_OPCODE();
@@ -5896 +5936 @@
 
     uint8_t bRm; IEM_OPCODE_GET_NEXT_U8(&bRm);
-    IEMOP_HLP_NO_LOCK_PREFIX(); /** @todo should probably not be raised until we've fetched all the opcode bytes? */
 
     /*
@@ -5903 +5942 @@
     if ((bRm & X86_MODRM_MOD_MASK) == (3 << X86_MODRM_MOD_SHIFT))
     {
+        IEMOP_HLP_DONE_DECODING_NO_LOCK_PREFIX();
         switch (pVCpu->iem.s.enmEffOpSize)
         {
@@ -5947 +5987 @@
                 IEM_MC_LOCAL(RTGCPTR, GCPtrEffDst);
                 IEM_MC_CALC_RM_EFF_ADDR(GCPtrEffDst, bRm, 0);
+                IEMOP_HLP_DONE_DECODING_NO_LOCK_PREFIX();
                 IEM_MC_FETCH_MEM_U8_ZX_U16(u16Value, pVCpu->iem.s.iEffSeg, GCPtrEffDst);
                 IEM_MC_STORE_GREG_U16(((bRm >> X86_MODRM_REG_SHIFT) & X86_MODRM_REG_SMASK) | pVCpu->iem.s.uRexReg, u16Value);
@@ -5958 +5999 @@
                 IEM_MC_LOCAL(RTGCPTR, GCPtrEffDst);
                 IEM_MC_CALC_RM_EFF_ADDR(GCPtrEffDst, bRm, 0);
+                IEMOP_HLP_DONE_DECODING_NO_LOCK_PREFIX();
                 IEM_MC_FETCH_MEM_U8_ZX_U32(u32Value, pVCpu->iem.s.iEffSeg, GCPtrEffDst);
                 IEM_MC_STORE_GREG_U32(((bRm >> X86_MODRM_REG_SHIFT) & X86_MODRM_REG_SMASK) | pVCpu->iem.s.uRexReg, u32Value);
@@ -5969 +6011 @@
                 IEM_MC_LOCAL(RTGCPTR, GCPtrEffDst);
                 IEM_MC_CALC_RM_EFF_ADDR(GCPtrEffDst, bRm, 0);
+                IEMOP_HLP_DONE_DECODING_NO_LOCK_PREFIX();
                 IEM_MC_FETCH_MEM_U8_ZX_U64(u64Value, pVCpu->iem.s.iEffSeg, GCPtrEffDst);
                 IEM_MC_STORE_GREG_U64(((bRm >> X86_MODRM_REG_SHIFT) & X86_MODRM_REG_SMASK) | pVCpu->iem.s.uRexReg, u64Value);
@@ -5988 +6031 @@
 
     uint8_t bRm; IEM_OPCODE_GET_NEXT_U8(&bRm);
-    IEMOP_HLP_NO_LOCK_PREFIX(); /** @todo should probably not be raised until we've fetched all the opcode bytes? */
 
     /** @todo Not entirely sure how the operand size prefix is handled here,
@@ -5998 +6040 @@
     if ((bRm & X86_MODRM_MOD_MASK) == (3 << X86_MODRM_MOD_SHIFT))
     {
+        IEMOP_HLP_DONE_DECODING_NO_LOCK_PREFIX();
         if (pVCpu->iem.s.enmEffOpSize != IEMMODE_64BIT)
         {
@@ -6028 +6071 @@
             IEM_MC_LOCAL(RTGCPTR, GCPtrEffDst);
             IEM_MC_CALC_RM_EFF_ADDR(GCPtrEffDst, bRm, 0);
+            IEMOP_HLP_DONE_DECODING_NO_LOCK_PREFIX();
             IEM_MC_FETCH_MEM_U16_ZX_U32(u32Value, pVCpu->iem.s.iEffSeg, GCPtrEffDst);
             IEM_MC_STORE_GREG_U32(((bRm >> X86_MODRM_REG_SHIFT) & X86_MODRM_REG_SMASK) | pVCpu->iem.s.uRexReg, u32Value);
@@ -6039 +6083 @@
             IEM_MC_LOCAL(RTGCPTR, GCPtrEffDst);
             IEM_MC_CALC_RM_EFF_ADDR(GCPtrEffDst, bRm, 0);
+            IEMOP_HLP_DONE_DECODING_NO_LOCK_PREFIX();
             IEM_MC_FETCH_MEM_U16_ZX_U64(u64Value, pVCpu->iem.s.iEffSeg, GCPtrEffDst);
             IEM_MC_STORE_GREG_U64(((bRm >> X86_MODRM_REG_SHIFT) & X86_MODRM_REG_SMASK) | pVCpu->iem.s.uRexReg, u64Value);
@@ -6083 +6128 @@
         /* register destination. */
         uint8_t u8Bit; IEM_OPCODE_GET_NEXT_U8(&u8Bit);
-        IEMOP_HLP_NO_LOCK_PREFIX();
+        IEMOP_HLP_DONE_DECODING_NO_LOCK_PREFIX();
 
         switch (pVCpu->iem.s.enmEffOpSize)
@@ -6141 +6186 @@
             fAccess = IEM_ACCESS_DATA_RW;
         else /* BT */
-        {
-            IEMOP_HLP_NO_LOCK_PREFIX();
             fAccess = IEM_ACCESS_DATA_R;
-        }
 
         /** @todo test negative bit offsets! */
@@ -6159 +6201 @@
                 uint8_t u8Bit; IEM_OPCODE_GET_NEXT_U8(&u8Bit);
                 IEM_MC_ASSIGN(u16Src, u8Bit & 0x0f);
+                if (pImpl->pfnLockedU16)
+                    IEMOP_HLP_DONE_DECODING();
+                else
+                    IEMOP_HLP_DONE_DECODING_NO_LOCK_PREFIX();
                 IEM_MC_FETCH_EFLAGS(EFlags);
                 IEM_MC_MEM_MAP(pu16Dst, fAccess, pVCpu->iem.s.iEffSeg, GCPtrEffDst, 0);
@@ -6182 +6228 @@
                 uint8_t u8Bit; IEM_OPCODE_GET_NEXT_U8(&u8Bit);
                 IEM_MC_ASSIGN(u32Src, u8Bit & 0x1f);
+                if (pImpl->pfnLockedU16)
+                    IEMOP_HLP_DONE_DECODING();
+                else
+                    IEMOP_HLP_DONE_DECODING_NO_LOCK_PREFIX();
                 IEM_MC_FETCH_EFLAGS(EFlags);
                 IEM_MC_MEM_MAP(pu32Dst, fAccess, pVCpu->iem.s.iEffSeg, GCPtrEffDst, 0);
@@ -6205 +6255 @@
                 uint8_t u8Bit; IEM_OPCODE_GET_NEXT_U8(&u8Bit);
                 IEM_MC_ASSIGN(u64Src, u8Bit & 0x3f);
+                if (pImpl->pfnLockedU16)
+                    IEMOP_HLP_DONE_DECODING();
+                else
+                    IEMOP_HLP_DONE_DECODING_NO_LOCK_PREFIX();
                 IEM_MC_FETCH_EFLAGS(EFlags);
                 IEM_MC_MEM_MAP(pu64Dst, fAccess, pVCpu->iem.s.iEffSeg, GCPtrEffDst, 0);
@@ -6261 +6315 @@
 
     uint8_t bRm; IEM_OPCODE_GET_NEXT_U8(&bRm);
-    IEMOP_HLP_NO_LOCK_PREFIX(); /** @todo should probably not be raised until we've fetched all the opcode bytes? */
 
     /*
@@ -6268 +6321 @@
     if ((bRm & X86_MODRM_MOD_MASK) == (3 << X86_MODRM_MOD_SHIFT))
     {
+        IEMOP_HLP_DONE_DECODING_NO_LOCK_PREFIX();
         switch (pVCpu->iem.s.enmEffOpSize)
         {
@@ -6312 +6366 @@
                 IEM_MC_LOCAL(RTGCPTR, GCPtrEffDst);
                 IEM_MC_CALC_RM_EFF_ADDR(GCPtrEffDst, bRm, 0);
+                IEMOP_HLP_DONE_DECODING_NO_LOCK_PREFIX();
                 IEM_MC_FETCH_MEM_U8_SX_U16(u16Value, pVCpu->iem.s.iEffSeg, GCPtrEffDst);
                 IEM_MC_STORE_GREG_U16(((bRm >> X86_MODRM_REG_SHIFT) & X86_MODRM_REG_SMASK) | pVCpu->iem.s.uRexReg, u16Value);
@@ -6323 +6378 @@
                 IEM_MC_LOCAL(RTGCPTR, GCPtrEffDst);
                 IEM_MC_CALC_RM_EFF_ADDR(GCPtrEffDst, bRm, 0);
+                IEMOP_HLP_DONE_DECODING_NO_LOCK_PREFIX();
                 IEM_MC_FETCH_MEM_U8_SX_U32(u32Value, pVCpu->iem.s.iEffSeg, GCPtrEffDst);
                 IEM_MC_STORE_GREG_U32(((bRm >> X86_MODRM_REG_SHIFT) & X86_MODRM_REG_SMASK) | pVCpu->iem.s.uRexReg, u32Value);
@@ -6334 +6390 @@
                 IEM_MC_LOCAL(RTGCPTR, GCPtrEffDst);
                 IEM_MC_CALC_RM_EFF_ADDR(GCPtrEffDst, bRm, 0);
+                IEMOP_HLP_DONE_DECODING_NO_LOCK_PREFIX();
                 IEM_MC_FETCH_MEM_U8_SX_U64(u64Value, pVCpu->iem.s.iEffSeg, GCPtrEffDst);
                 IEM_MC_STORE_GREG_U64(((bRm >> X86_MODRM_REG_SHIFT) & X86_MODRM_REG_SMASK) | pVCpu->iem.s.uRexReg, u64Value);
@@ -6353 +6410 @@
 
     uint8_t bRm; IEM_OPCODE_GET_NEXT_U8(&bRm);
-    IEMOP_HLP_NO_LOCK_PREFIX(); /** @todo should probably not be raised until we've fetched all the opcode bytes? */
 
     /** @todo Not entirely sure how the operand size prefix is handled here,
@@ -6363 +6419 @@
     if ((bRm & X86_MODRM_MOD_MASK) == (3 << X86_MODRM_MOD_SHIFT))
     {
+        IEMOP_HLP_DONE_DECODING_NO_LOCK_PREFIX();
         if (pVCpu->iem.s.enmEffOpSize != IEMMODE_64BIT)
         {
@@ -6393 +6450 @@
             IEM_MC_LOCAL(RTGCPTR, GCPtrEffDst);
             IEM_MC_CALC_RM_EFF_ADDR(GCPtrEffDst, bRm, 0);
+            IEMOP_HLP_DONE_DECODING_NO_LOCK_PREFIX();
             IEM_MC_FETCH_MEM_U16_SX_U32(u32Value, pVCpu->iem.s.iEffSeg, GCPtrEffDst);
             IEM_MC_STORE_GREG_U32(((bRm >> X86_MODRM_REG_SHIFT) & X86_MODRM_REG_SMASK) | pVCpu->iem.s.uRexReg, u32Value);
@@ -6404 +6462 @@
             IEM_MC_LOCAL(RTGCPTR, GCPtrEffDst);
             IEM_MC_CALC_RM_EFF_ADDR(GCPtrEffDst, bRm, 0);
+            IEMOP_HLP_DONE_DECODING_NO_LOCK_PREFIX();
             IEM_MC_FETCH_MEM_U16_SX_U64(u64Value, pVCpu->iem.s.iEffSeg, GCPtrEffDst);
             IEM_MC_STORE_GREG_U64(((bRm >> X86_MODRM_REG_SHIFT) & X86_MODRM_REG_SMASK) | pVCpu->iem.s.uRexReg, u64Value);
@@ -6426 +6485 @@
     if ((bRm & X86_MODRM_MOD_MASK) == (3 << X86_MODRM_MOD_SHIFT))
     {
-        IEMOP_HLP_NO_LOCK_PREFIX();
+        IEMOP_HLP_DONE_DECODING_NO_LOCK_PREFIX();
 
         IEM_MC_BEGIN(3, 0);
@@ -6486 +6545 @@
     if ((bRm & X86_MODRM_MOD_MASK) == (3 << X86_MODRM_MOD_SHIFT))
     {
-        IEMOP_HLP_NO_LOCK_PREFIX();
+        IEMOP_HLP_DONE_DECODING_NO_LOCK_PREFIX();
 
         switch (pVCpu->iem.s.enmEffOpSize)
@@ -6811 +6870 @@
 FNIEMOP_DEF_1(iemOpCommonBswapGReg, uint8_t, iReg)
 {
-    IEMOP_HLP_NO_LOCK_PREFIX();
+    IEMOP_HLP_DONE_DECODING_NO_LOCK_PREFIX();
     switch (pVCpu->iem.s.enmEffOpSize)
     {
@@ -7466 +7525 @@
     IEMOP_MNEMONIC("pop es");
     IEMOP_HLP_NO_64BIT();
-    IEMOP_HLP_NO_LOCK_PREFIX();
+    IEMOP_HLP_DONE_DECODING_NO_LOCK_PREFIX();
     return IEM_MC_DEFER_TO_CIMPL_2(iemCImpl_pop_Sreg, X86_SREG_ES, pVCpu->iem.s.enmEffOpSize);
 }
@@ -7602 +7661 @@
 {
     IEMOP_MNEMONIC("pop ss"); /** @todo implies instruction fusing? */
-    IEMOP_HLP_NO_LOCK_PREFIX();
+    IEMOP_HLP_DONE_DECODING_NO_LOCK_PREFIX();
     IEMOP_HLP_NO_64BIT();
     return IEM_MC_DEFER_TO_CIMPL_2(iemCImpl_pop_Sreg, X86_SREG_SS, pVCpu->iem.s.enmEffOpSize);
@@ -7668 +7727 @@
 {
     IEMOP_MNEMONIC("pop ds");
-    IEMOP_HLP_NO_LOCK_PREFIX();
+    IEMOP_HLP_DONE_DECODING_NO_LOCK_PREFIX();
     IEMOP_HLP_NO_64BIT();
     return IEM_MC_DEFER_TO_CIMPL_2(iemCImpl_pop_Sreg, X86_SREG_DS, pVCpu->iem.s.enmEffOpSize);
@@ -7896 +7955 @@
 {
     IEMOP_MNEMONIC("cmp Eb,Gb");
-    IEMOP_HLP_NO_LOCK_PREFIX(); /** @todo do we have to decode the whole instruction first?  */
     return FNIEMOP_CALL_1(iemOpHlpBinaryOperator_rm_r8, &g_iemAImpl_cmp);
 }
@@ -7905 +7963 @@
 {
     IEMOP_MNEMONIC("cmp Ev,Gv");
-    IEMOP_HLP_NO_LOCK_PREFIX(); /** @todo do we have to decode the whole instruction first?  */
     return FNIEMOP_CALL_1(iemOpHlpBinaryOperator_rm_rv, &g_iemAImpl_cmp);
 }
@@ -7962 +8019 @@
 FNIEMOP_DEF_2(iemOpCommonUnaryGReg, PCIEMOPUNARYSIZES, pImpl, uint8_t, iReg)
 {
-    IEMOP_HLP_NO_LOCK_PREFIX();
+    IEMOP_HLP_DONE_DECODING_NO_LOCK_PREFIX();
     switch (pVCpu->iem.s.enmEffOpSize)
     {
@@ -8361 +8418 @@
 FNIEMOP_DEF_1(iemOpCommonPushGReg, uint8_t, iReg)
 {
-    IEMOP_HLP_NO_LOCK_PREFIX();
+    IEMOP_HLP_DONE_DECODING_NO_LOCK_PREFIX();
     if (pVCpu->iem.s.enmCpuMode == IEMMODE_64BIT)
     {
@@ -8482 +8539 @@
 FNIEMOP_DEF_1(iemOpCommonPopGReg, uint8_t, iReg)
 {
-    IEMOP_HLP_NO_LOCK_PREFIX();
+    IEMOP_HLP_DONE_DECODING_NO_LOCK_PREFIX();
     if (pVCpu->iem.s.enmCpuMode == IEMMODE_64BIT)
     {
@@ -8828 +8885 @@
         {
             uint16_t u16Imm; IEM_OPCODE_GET_NEXT_U16(&u16Imm);
-            IEMOP_HLP_NO_LOCK_PREFIX();
+            IEMOP_HLP_DONE_DECODING_NO_LOCK_PREFIX();
             IEM_MC_BEGIN(0,0);
             IEM_MC_PUSH_U16(u16Imm);
@@ -8839 +8896 @@
         {
             uint32_t u32Imm; IEM_OPCODE_GET_NEXT_U32(&u32Imm);
-            IEMOP_HLP_NO_LOCK_PREFIX();
+            IEMOP_HLP_DONE_DECODING_NO_LOCK_PREFIX();
             IEM_MC_BEGIN(0,0);
             IEM_MC_PUSH_U32(u32Imm);
@@ -8850 +8907 @@
         {
             uint64_t u64Imm; IEM_OPCODE_GET_NEXT_S32_SX_U64(&u64Imm);
-            IEMOP_HLP_NO_LOCK_PREFIX();
+            IEMOP_HLP_DONE_DECODING_NO_LOCK_PREFIX();
             IEM_MC_BEGIN(0,0);
             IEM_MC_PUSH_U64(u64Imm);
@@ -9030 +9087 @@
     IEMOP_HLP_MIN_186();
     int8_t i8Imm; IEM_OPCODE_GET_NEXT_S8(&i8Imm);
-    IEMOP_HLP_NO_LOCK_PREFIX();
+    IEMOP_HLP_DONE_DECODING_NO_LOCK_PREFIX();
     IEMOP_HLP_DEFAULT_64BIT_OP_SIZE();
 
@@ -9211 +9268 @@
 {
     IEMOP_HLP_MIN_186();
-    IEMOP_HLP_NO_LOCK_PREFIX();
+    IEMOP_HLP_DONE_DECODING_NO_LOCK_PREFIX();
     if (pVCpu->iem.s.fPrefixes & (IEM_OP_PRF_REPNZ | IEM_OP_PRF_REPZ))
     {
@@ -9241 +9298 @@
 {
     IEMOP_HLP_MIN_186();
-    IEMOP_HLP_NO_LOCK_PREFIX();
+    IEMOP_HLP_DONE_DECODING_NO_LOCK_PREFIX();
     if (pVCpu->iem.s.fPrefixes & (IEM_OP_PRF_REPZ | IEM_OP_PRF_REPNZ))
     {
@@ -9303 +9360 @@
 {
     IEMOP_HLP_MIN_186();
-    IEMOP_HLP_NO_LOCK_PREFIX();
+    IEMOP_HLP_DONE_DECODING_NO_LOCK_PREFIX();
     if (pVCpu->iem.s.fPrefixes & (IEM_OP_PRF_REPNZ | IEM_OP_PRF_REPZ))
     {
@@ -9333 +9390 @@
 {
     IEMOP_HLP_MIN_186();
-    IEMOP_HLP_NO_LOCK_PREFIX();
+    IEMOP_HLP_DONE_DECODING_NO_LOCK_PREFIX();
     if (pVCpu->iem.s.fPrefixes & (IEM_OP_PRF_REPZ | IEM_OP_PRF_REPNZ))
     {
@@ -9396 +9453 @@
     IEMOP_MNEMONIC("jo  Jb");
     int8_t i8Imm; IEM_OPCODE_GET_NEXT_S8(&i8Imm);
-    IEMOP_HLP_NO_LOCK_PREFIX();
+    IEMOP_HLP_DONE_DECODING_NO_LOCK_PREFIX();
     IEMOP_HLP_DEFAULT_64BIT_OP_SIZE();
 
@@ -9415 +9472 @@
     IEMOP_MNEMONIC("jno Jb");
     int8_t i8Imm; IEM_OPCODE_GET_NEXT_S8(&i8Imm);
-    IEMOP_HLP_NO_LOCK_PREFIX();
+    IEMOP_HLP_DONE_DECODING_NO_LOCK_PREFIX();
     IEMOP_HLP_DEFAULT_64BIT_OP_SIZE();
 
@@ -9433 +9490 @@
     IEMOP_MNEMONIC("jc/jnae Jb");
     int8_t i8Imm; IEM_OPCODE_GET_NEXT_S8(&i8Imm);
-    IEMOP_HLP_NO_LOCK_PREFIX();
+    IEMOP_HLP_DONE_DECODING_NO_LOCK_PREFIX();
     IEMOP_HLP_DEFAULT_64BIT_OP_SIZE();
 
@@ -9452 +9509 @@
     IEMOP_MNEMONIC("jnc/jnb Jb");
     int8_t i8Imm; IEM_OPCODE_GET_NEXT_S8(&i8Imm);
-    IEMOP_HLP_NO_LOCK_PREFIX();
+    IEMOP_HLP_DONE_DECODING_NO_LOCK_PREFIX();
     IEMOP_HLP_DEFAULT_64BIT_OP_SIZE();
 
@@ -9471 +9528 @@
     IEMOP_MNEMONIC("je/jz   Jb");
     int8_t i8Imm; IEM_OPCODE_GET_NEXT_S8(&i8Imm);
-    IEMOP_HLP_NO_LOCK_PREFIX();
+    IEMOP_HLP_DONE_DECODING_NO_LOCK_PREFIX();
     IEMOP_HLP_DEFAULT_64BIT_OP_SIZE();
 
@@ -9490 +9547 @@
     IEMOP_MNEMONIC("jne/jnz Jb");
     int8_t i8Imm; IEM_OPCODE_GET_NEXT_S8(&i8Imm);
-    IEMOP_HLP_NO_LOCK_PREFIX();
+    IEMOP_HLP_DONE_DECODING_NO_LOCK_PREFIX();
     IEMOP_HLP_DEFAULT_64BIT_OP_SIZE();
 
@@ -9509 +9566 @@
     IEMOP_MNEMONIC("jbe/jna Jb");
     int8_t i8Imm; IEM_OPCODE_GET_NEXT_S8(&i8Imm);
-    IEMOP_HLP_NO_LOCK_PREFIX();
+    IEMOP_HLP_DONE_DECODING_NO_LOCK_PREFIX();
     IEMOP_HLP_DEFAULT_64BIT_OP_SIZE();
 
@@ -9528 +9585 @@
     IEMOP_MNEMONIC("jnbe/ja Jb");
     int8_t i8Imm; IEM_OPCODE_GET_NEXT_S8(&i8Imm);
-    IEMOP_HLP_NO_LOCK_PREFIX();
+    IEMOP_HLP_DONE_DECODING_NO_LOCK_PREFIX();
     IEMOP_HLP_DEFAULT_64BIT_OP_SIZE();
 
@@ -9547 +9604 @@
     IEMOP_MNEMONIC("js  Jb");
     int8_t i8Imm; IEM_OPCODE_GET_NEXT_S8(&i8Imm);
-    IEMOP_HLP_NO_LOCK_PREFIX();
+    IEMOP_HLP_DONE_DECODING_NO_LOCK_PREFIX();
     IEMOP_HLP_DEFAULT_64BIT_OP_SIZE();
 
@@ -9566 +9623 @@
     IEMOP_MNEMONIC("jns Jb");
     int8_t i8Imm; IEM_OPCODE_GET_NEXT_S8(&i8Imm);
-    IEMOP_HLP_NO_LOCK_PREFIX();
+    IEMOP_HLP_DONE_DECODING_NO_LOCK_PREFIX();
     IEMOP_HLP_DEFAULT_64BIT_OP_SIZE();
 
@@ -9585 +9642 @@
     IEMOP_MNEMONIC("jp  Jb");
     int8_t i8Imm; IEM_OPCODE_GET_NEXT_S8(&i8Imm);
-    IEMOP_HLP_NO_LOCK_PREFIX();
+    IEMOP_HLP_DONE_DECODING_NO_LOCK_PREFIX();
     IEMOP_HLP_DEFAULT_64BIT_OP_SIZE();
 
@@ -9604 +9661 @@
     IEMOP_MNEMONIC("jnp Jb");
     int8_t i8Imm; IEM_OPCODE_GET_NEXT_S8(&i8Imm);
-    IEMOP_HLP_NO_LOCK_PREFIX();
+    IEMOP_HLP_DONE_DECODING_NO_LOCK_PREFIX();
     IEMOP_HLP_DEFAULT_64BIT_OP_SIZE();
 
@@ -9623 +9680 @@
     IEMOP_MNEMONIC("jl/jnge Jb");
     int8_t i8Imm; IEM_OPCODE_GET_NEXT_S8(&i8Imm);
-    IEMOP_HLP_NO_LOCK_PREFIX();
+    IEMOP_HLP_DONE_DECODING_NO_LOCK_PREFIX();
     IEMOP_HLP_DEFAULT_64BIT_OP_SIZE();
 
@@ -9642 +9699 @@
     IEMOP_MNEMONIC("jnl/jge Jb");
     int8_t i8Imm; IEM_OPCODE_GET_NEXT_S8(&i8Imm);
-    IEMOP_HLP_NO_LOCK_PREFIX();
+    IEMOP_HLP_DONE_DECODING_NO_LOCK_PREFIX();
     IEMOP_HLP_DEFAULT_64BIT_OP_SIZE();
 
@@ -9661 +9718 @@
     IEMOP_MNEMONIC("jle/jng Jb");
     int8_t i8Imm; IEM_OPCODE_GET_NEXT_S8(&i8Imm);
-    IEMOP_HLP_NO_LOCK_PREFIX();
+    IEMOP_HLP_DONE_DECODING_NO_LOCK_PREFIX();
     IEMOP_HLP_DEFAULT_64BIT_OP_SIZE();
 
@@ -9680 +9737 @@
     IEMOP_MNEMONIC("jnle/jg Jb");
     int8_t i8Imm; IEM_OPCODE_GET_NEXT_S8(&i8Imm);
-    IEMOP_HLP_NO_LOCK_PREFIX();
+    IEMOP_HLP_DONE_DECODING_NO_LOCK_PREFIX();
     IEMOP_HLP_DEFAULT_64BIT_OP_SIZE();
 
@@ -9705 +9762 @@
         /* register target */
         uint8_t u8Imm; IEM_OPCODE_GET_NEXT_U8(&u8Imm);
-        IEMOP_HLP_NO_LOCK_PREFIX();
+        IEMOP_HLP_DONE_DECODING_NO_LOCK_PREFIX();
         IEM_MC_BEGIN(3, 0);
         IEM_MC_ARG(uint8_t *,       pu8Dst,                 0);
@@ -9724 +9781 @@
         if (pImpl->pfnLockedU8)
             fAccess = IEM_ACCESS_DATA_RW;
-        else
-        {   /* CMP */
-            IEMOP_HLP_NO_LOCK_PREFIX();
+        else /* CMP */
             fAccess = IEM_ACCESS_DATA_R;
-        }
         IEM_MC_BEGIN(3, 2);
         IEM_MC_ARG(uint8_t *,       pu8Dst,                 0);
@@ -9737 +9791 @@
         uint8_t u8Imm; IEM_OPCODE_GET_NEXT_U8(&u8Imm);
         IEM_MC_ARG_CONST(uint8_t,   u8Src, /*=*/ u8Imm,     1);
+        if (pImpl->pfnLockedU8)
+            IEMOP_HLP_DONE_DECODING();
+        else
+            IEMOP_HLP_DONE_DECODING_NO_LOCK_PREFIX();
 
         IEM_MC_MEM_MAP(pu8Dst, fAccess, pVCpu->iem.s.iEffSeg, GCPtrEffDst, 0 /*arg*/);
@@ -9769 +9827 @@
                 /* register target */
                 uint16_t u16Imm; IEM_OPCODE_GET_NEXT_U16(&u16Imm);
-                IEMOP_HLP_NO_LOCK_PREFIX();
+                IEMOP_HLP_DONE_DECODING_NO_LOCK_PREFIX();
                 IEM_MC_BEGIN(3, 0);
                 IEM_MC_ARG(uint16_t *,      pu16Dst,                0);
@@ -9788 +9846 @@
                 if (pImpl->pfnLockedU16)
                     fAccess = IEM_ACCESS_DATA_RW;
-                else
-                {   /* CMP, TEST */
-                    IEMOP_HLP_NO_LOCK_PREFIX();
+                else /* CMP, TEST */
                     fAccess = IEM_ACCESS_DATA_R;
-                }
                 IEM_MC_BEGIN(3, 2);
                 IEM_MC_ARG(uint16_t *,      pu16Dst,                0);
@@ -9802 +9857 @@
                 uint16_t u16Imm; IEM_OPCODE_GET_NEXT_U16(&u16Imm);
                 IEM_MC_ASSIGN(u16Src, u16Imm);
+                if (pImpl->pfnLockedU16)
+                    IEMOP_HLP_DONE_DECODING();
+                else
+                    IEMOP_HLP_DONE_DECODING_NO_LOCK_PREFIX();
                 IEM_MC_MEM_MAP(pu16Dst, fAccess, pVCpu->iem.s.iEffSeg, GCPtrEffDst, 0 /*arg*/);
                 IEM_MC_FETCH_EFLAGS(EFlags);
@@ -9823 +9882 @@
                 /* register target */
                 uint32_t u32Imm; IEM_OPCODE_GET_NEXT_U32(&u32Imm);
-                IEMOP_HLP_NO_LOCK_PREFIX();
+                IEMOP_HLP_DONE_DECODING_NO_LOCK_PREFIX();
                 IEM_MC_BEGIN(3, 0);
                 IEM_MC_ARG(uint32_t *,      pu32Dst,                0);
@@ -9843 +9902 @@
                 if (pImpl->pfnLockedU32)
                     fAccess = IEM_ACCESS_DATA_RW;
-                else
-                {   /* CMP, TEST */
-                    IEMOP_HLP_NO_LOCK_PREFIX();
+                else /* CMP, TEST */
                     fAccess = IEM_ACCESS_DATA_R;
-                }
                 IEM_MC_BEGIN(3, 2);
                 IEM_MC_ARG(uint32_t *,      pu32Dst,                0);
@@ -9857 +9913 @@
                 uint32_t u32Imm; IEM_OPCODE_GET_NEXT_U32(&u32Imm);
                 IEM_MC_ASSIGN(u32Src, u32Imm);
+                if (pImpl->pfnLockedU32)
+                    IEMOP_HLP_DONE_DECODING();
+                else
+                    IEMOP_HLP_DONE_DECODING_NO_LOCK_PREFIX();
                 IEM_MC_MEM_MAP(pu32Dst, fAccess, pVCpu->iem.s.iEffSeg, GCPtrEffDst, 0 /*arg*/);
                 IEM_MC_FETCH_EFLAGS(EFlags);
@@ -9878 +9938 @@
                 /* register target */
                 uint64_t u64Imm; IEM_OPCODE_GET_NEXT_S32_SX_U64(&u64Imm);
-                IEMOP_HLP_NO_LOCK_PREFIX();
+                IEMOP_HLP_DONE_DECODING_NO_LOCK_PREFIX();
                 IEM_MC_BEGIN(3, 0);
                 IEM_MC_ARG(uint64_t *,      pu64Dst,                0);
@@ -9897 +9957 @@
                 if (pImpl->pfnLockedU64)
                     fAccess = IEM_ACCESS_DATA_RW;
-                else
-                {   /* CMP */
-                    IEMOP_HLP_NO_LOCK_PREFIX();
+                else /* CMP */
                     fAccess = IEM_ACCESS_DATA_R;
-                }
                 IEM_MC_BEGIN(3, 2);
                 IEM_MC_ARG(uint64_t *,      pu64Dst,                0);
@@ -9910 +9967 @@
                 IEM_MC_CALC_RM_EFF_ADDR(GCPtrEffDst, bRm, 4);
                 uint64_t u64Imm; IEM_OPCODE_GET_NEXT_S32_SX_U64(&u64Imm);
+                if (pImpl->pfnLockedU64)
+                    IEMOP_HLP_DONE_DECODING();
+                else
+                    IEMOP_HLP_DONE_DECODING_NO_LOCK_PREFIX();
                 IEM_MC_ASSIGN(u64Src, u64Imm);
                 IEM_MC_MEM_MAP(pu64Dst, fAccess, pVCpu->iem.s.iEffSeg, GCPtrEffDst, 0 /*arg*/);
@@ -9953 +10014 @@
          * Register target
          */
-        IEMOP_HLP_NO_LOCK_PREFIX();
+        IEMOP_HLP_DONE_DECODING_NO_LOCK_PREFIX();
         uint8_t u8Imm; IEM_OPCODE_GET_NEXT_U8(&u8Imm);
         switch (pVCpu->iem.s.enmEffOpSize)
@@ -10015 +10076 @@
         if (pImpl->pfnLockedU16)
             fAccess = IEM_ACCESS_DATA_RW;
-        else
-        {   /* CMP */
-            IEMOP_HLP_NO_LOCK_PREFIX();
+        else /* CMP */
             fAccess = IEM_ACCESS_DATA_R;
-        }
 
         switch (pVCpu->iem.s.enmEffOpSize)
@@ -10034 +10092 @@
                 uint8_t u8Imm; IEM_OPCODE_GET_NEXT_U8(&u8Imm);
                 IEM_MC_ASSIGN(u16Src, (int8_t)u8Imm);
+                if (pImpl->pfnLockedU16)
+                    IEMOP_HLP_DONE_DECODING();
+                else
+                    IEMOP_HLP_DONE_DECODING_NO_LOCK_PREFIX();
                 IEM_MC_MEM_MAP(pu16Dst, fAccess, pVCpu->iem.s.iEffSeg, GCPtrEffDst, 0 /*arg*/);
                 IEM_MC_FETCH_EFLAGS(EFlags);
@@ -10059 +10121 @@
                 uint8_t u8Imm; IEM_OPCODE_GET_NEXT_U8(&u8Imm);
                 IEM_MC_ASSIGN(u32Src, (int8_t)u8Imm);
+                if (pImpl->pfnLockedU32)
+                    IEMOP_HLP_DONE_DECODING();
+                else
+                    IEMOP_HLP_DONE_DECODING_NO_LOCK_PREFIX();
                 IEM_MC_MEM_MAP(pu32Dst, fAccess, pVCpu->iem.s.iEffSeg, GCPtrEffDst, 0 /*arg*/);
                 IEM_MC_FETCH_EFLAGS(EFlags);
@@ -10084 +10150 @@
                 uint8_t u8Imm; IEM_OPCODE_GET_NEXT_U8(&u8Imm);
                 IEM_MC_ASSIGN(u64Src, (int8_t)u8Imm);
+                if (pImpl->pfnLockedU64)
+                    IEMOP_HLP_DONE_DECODING();
+                else
+                    IEMOP_HLP_DONE_DECODING_NO_LOCK_PREFIX();
                 IEM_MC_MEM_MAP(pu64Dst, fAccess, pVCpu->iem.s.iEffSeg, GCPtrEffDst, 0 /*arg*/);
                 IEM_MC_FETCH_EFLAGS(EFlags);
@@ -10107 +10177 @@
 {
     IEMOP_MNEMONIC("test Eb,Gb");
-    IEMOP_HLP_NO_LOCK_PREFIX(); /** @todo do we have to decode the whole instruction first?  */
     IEMOP_VERIFICATION_UNDEFINED_EFLAGS(X86_EFL_AF);
     return FNIEMOP_CALL_1(iemOpHlpBinaryOperator_rm_r8, &g_iemAImpl_test);
@@ -10117 +10186 @@
 {
     IEMOP_MNEMONIC("test Ev,Gv");
-    IEMOP_HLP_NO_LOCK_PREFIX(); /** @todo do we have to decode the whole instruction first?  */
     IEMOP_VERIFICATION_UNDEFINED_EFLAGS(X86_EFL_AF);
     return FNIEMOP_CALL_1(iemOpHlpBinaryOperator_rm_rv, &g_iemAImpl_test);
@@ -10134 +10202 @@
     if ((bRm & X86_MODRM_MOD_MASK) == (3 << X86_MODRM_MOD_SHIFT))
     {
-        IEMOP_HLP_NO_LOCK_PREFIX();
+        IEMOP_HLP_DONE_DECODING_NO_LOCK_PREFIX();
 
         IEM_MC_BEGIN(0, 2);
@@ -10183 +10251 @@
     if ((bRm & X86_MODRM_MOD_MASK) == (3 << X86_MODRM_MOD_SHIFT))
     {
-        IEMOP_HLP_NO_LOCK_PREFIX();
+        IEMOP_HLP_DONE_DECODING_NO_LOCK_PREFIX();
 
         switch (pVCpu->iem.s.enmEffOpSize)
@@ -10302 +10370 @@
     uint8_t bRm;
     IEM_OPCODE_GET_NEXT_U8(&bRm);
-    IEMOP_HLP_NO_LOCK_PREFIX(); /** @todo should probably not be raised until we've fetched all the opcode bytes? */
 
     /*
@@ -10309 +10376 @@
     if ((bRm & X86_MODRM_MOD_MASK) == (3 << X86_MODRM_MOD_SHIFT))
     {
+        IEMOP_HLP_DONE_DECODING_NO_LOCK_PREFIX();
         IEM_MC_BEGIN(0, 1);
         IEM_MC_LOCAL(uint8_t, u8Value);
@@ -10325 +10393 @@
         IEM_MC_LOCAL(RTGCPTR, GCPtrEffDst);
         IEM_MC_CALC_RM_EFF_ADDR(GCPtrEffDst, bRm, 0);
+        IEMOP_HLP_DONE_DECODING_NO_LOCK_PREFIX();
         IEM_MC_FETCH_GREG_U8(u8Value, ((bRm >> X86_MODRM_REG_SHIFT) & X86_MODRM_REG_SMASK) | pVCpu->iem.s.uRexReg);
         IEM_MC_STORE_MEM_U8(pVCpu->iem.s.iEffSeg, GCPtrEffDst, u8Value);
@@ -10341 +10410 @@
 
     uint8_t bRm; IEM_OPCODE_GET_NEXT_U8(&bRm);
-    IEMOP_HLP_NO_LOCK_PREFIX(); /** @todo should probably not be raised until we've fetched all the opcode bytes? */
 
     /*
@@ -10348 +10416 @@
     if ((bRm & X86_MODRM_MOD_MASK) == (3 << X86_MODRM_MOD_SHIFT))
     {
+        IEMOP_HLP_DONE_DECODING_NO_LOCK_PREFIX();
         switch (pVCpu->iem.s.enmEffOpSize)
         {
@@ -10390 +10459 @@
                 IEM_MC_LOCAL(RTGCPTR, GCPtrEffDst);
                 IEM_MC_CALC_RM_EFF_ADDR(GCPtrEffDst, bRm, 0);
+                IEMOP_HLP_DONE_DECODING_NO_LOCK_PREFIX();
                 IEM_MC_FETCH_GREG_U16(u16Value, ((bRm >> X86_MODRM_REG_SHIFT) & X86_MODRM_REG_SMASK) | pVCpu->iem.s.uRexReg);
                 IEM_MC_STORE_MEM_U16(pVCpu->iem.s.iEffSeg, GCPtrEffDst, u16Value);
@@ -10401 +10471 @@
                 IEM_MC_LOCAL(RTGCPTR, GCPtrEffDst);
                 IEM_MC_CALC_RM_EFF_ADDR(GCPtrEffDst, bRm, 0);
+                IEMOP_HLP_DONE_DECODING_NO_LOCK_PREFIX();
                 IEM_MC_FETCH_GREG_U32(u32Value, ((bRm >> X86_MODRM_REG_SHIFT) & X86_MODRM_REG_SMASK) | pVCpu->iem.s.uRexReg);
                 IEM_MC_STORE_MEM_U32(pVCpu->iem.s.iEffSeg, GCPtrEffDst, u32Value);
@@ -10412 +10483 @@
                 IEM_MC_LOCAL(RTGCPTR, GCPtrEffDst);
                 IEM_MC_CALC_RM_EFF_ADDR(GCPtrEffDst, bRm, 0);
+                IEMOP_HLP_DONE_DECODING_NO_LOCK_PREFIX();
                 IEM_MC_FETCH_GREG_U64(u64Value, ((bRm >> X86_MODRM_REG_SHIFT) & X86_MODRM_REG_SMASK) | pVCpu->iem.s.uRexReg);
                 IEM_MC_STORE_MEM_U64(pVCpu->iem.s.iEffSeg, GCPtrEffDst, u64Value);
@@ -10429 +10501 @@
 
     uint8_t bRm; IEM_OPCODE_GET_NEXT_U8(&bRm);
-    IEMOP_HLP_NO_LOCK_PREFIX(); /** @todo should probably not be raised until we've fetched all the opcode bytes? */
 
     /*
@@ -10436 +10507 @@
     if ((bRm & X86_MODRM_MOD_MASK) == (3 << X86_MODRM_MOD_SHIFT))
     {
+        IEMOP_HLP_DONE_DECODING_NO_LOCK_PREFIX();
         IEM_MC_BEGIN(0, 1);
         IEM_MC_LOCAL(uint8_t, u8Value);
@@ -10452 +10524 @@
         IEM_MC_LOCAL(RTGCPTR, GCPtrEffDst);
         IEM_MC_CALC_RM_EFF_ADDR(GCPtrEffDst, bRm, 0);
+        IEMOP_HLP_DONE_DECODING_NO_LOCK_PREFIX();
         IEM_MC_FETCH_MEM_U8(u8Value, pVCpu->iem.s.iEffSeg, GCPtrEffDst);
         IEM_MC_STORE_GREG_U8(((bRm >> X86_MODRM_REG_SHIFT) & X86_MODRM_REG_SMASK) | pVCpu->iem.s.uRexReg, u8Value);
@@ -10467 +10540 @@
 
     uint8_t bRm; IEM_OPCODE_GET_NEXT_U8(&bRm);
-    IEMOP_HLP_NO_LOCK_PREFIX(); /** @todo should probably not be raised until we've fetched all the opcode bytes? */
 
     /*
@@ -10474 +10546 @@
     if ((bRm & X86_MODRM_MOD_MASK) == (3 << X86_MODRM_MOD_SHIFT))
     {
+        IEMOP_HLP_DONE_DECODING_NO_LOCK_PREFIX();
         switch (pVCpu->iem.s.enmEffOpSize)
         {
@@ -10516 +10589 @@
                 IEM_MC_LOCAL(RTGCPTR, GCPtrEffDst);
                 IEM_MC_CALC_RM_EFF_ADDR(GCPtrEffDst, bRm, 0);
+                IEMOP_HLP_DONE_DECODING_NO_LOCK_PREFIX();
                 IEM_MC_FETCH_MEM_U16(u16Value, pVCpu->iem.s.iEffSeg, GCPtrEffDst);
                 IEM_MC_STORE_GREG_U16(((bRm >> X86_MODRM_REG_SHIFT) & X86_MODRM_REG_SMASK) | pVCpu->iem.s.uRexReg, u16Value);
@@ -10527 +10601 @@
                 IEM_MC_LOCAL(RTGCPTR, GCPtrEffDst);
                 IEM_MC_CALC_RM_EFF_ADDR(GCPtrEffDst, bRm, 0);
+                IEMOP_HLP_DONE_DECODING_NO_LOCK_PREFIX();
                 IEM_MC_FETCH_MEM_U32(u32Value, pVCpu->iem.s.iEffSeg, GCPtrEffDst);
                 IEM_MC_STORE_GREG_U32(((bRm >> X86_MODRM_REG_SHIFT) & X86_MODRM_REG_SMASK) | pVCpu->iem.s.uRexReg, u32Value);
@@ -10538 +10613 @@
                 IEM_MC_LOCAL(RTGCPTR, GCPtrEffDst);
                 IEM_MC_CALC_RM_EFF_ADDR(GCPtrEffDst, bRm, 0);
+                IEMOP_HLP_DONE_DECODING_NO_LOCK_PREFIX();
                 IEM_MC_FETCH_MEM_U64(u64Value, pVCpu->iem.s.iEffSeg, GCPtrEffDst);
                 IEM_MC_STORE_GREG_U64(((bRm >> X86_MODRM_REG_SHIFT) & X86_MODRM_REG_SMASK) | pVCpu->iem.s.uRexReg, u64Value);
@@ -10566 +10642 @@
 
     uint8_t bRm; IEM_OPCODE_GET_NEXT_U8(&bRm);
-    IEMOP_HLP_NO_LOCK_PREFIX(); /** @todo should probably not be raised until we've fetched all the opcode bytes? */
 
     /*
@@ -10582 +10657 @@
     if ((bRm & X86_MODRM_MOD_MASK) == (3 << X86_MODRM_MOD_SHIFT))
     {
+        IEMOP_HLP_DONE_DECODING_NO_LOCK_PREFIX();
         switch (pVCpu->iem.s.enmEffOpSize)
         {
@@ -10625 +10701 @@
         IEM_MC_LOCAL(RTGCPTR,   GCPtrEffDst);
         IEM_MC_CALC_RM_EFF_ADDR(GCPtrEffDst, bRm, 0);
+        IEMOP_HLP_DONE_DECODING_NO_LOCK_PREFIX();
         IEM_MC_FETCH_SREG_U16(u16Value, iSegReg);
         IEM_MC_STORE_MEM_U16(pVCpu->iem.s.iEffSeg, GCPtrEffDst, u16Value);
@@ -10641 +10718 @@
     IEMOP_MNEMONIC("lea Gv,M");
     uint8_t bRm; IEM_OPCODE_GET_NEXT_U8(&bRm);
-    IEMOP_HLP_NO_LOCK_PREFIX(); /** @todo should probably not be raised until we've fetched all the opcode bytes? */
     if ((bRm & X86_MODRM_MOD_MASK) == (3 << X86_MODRM_MOD_SHIFT))
         return IEMOP_RAISE_INVALID_OPCODE(); /* no register form */
@@ -10652 +10728 @@
             IEM_MC_LOCAL(uint16_t, u16Cast);
             IEM_MC_CALC_RM_EFF_ADDR(GCPtrEffSrc, bRm, 0);
+            IEMOP_HLP_DONE_DECODING_NO_LOCK_PREFIX();
             IEM_MC_ASSIGN_TO_SMALLER(u16Cast, GCPtrEffSrc);
             IEM_MC_STORE_GREG_U16(((bRm >> X86_MODRM_REG_SHIFT) & X86_MODRM_REG_SMASK) | pVCpu->iem.s.uRexReg, u16Cast);
@@ -10663 +10740 @@
             IEM_MC_LOCAL(uint32_t, u32Cast);
             IEM_MC_CALC_RM_EFF_ADDR(GCPtrEffSrc, bRm, 0);
+            IEMOP_HLP_DONE_DECODING_NO_LOCK_PREFIX();
             IEM_MC_ASSIGN_TO_SMALLER(u32Cast, GCPtrEffSrc);
             IEM_MC_STORE_GREG_U32(((bRm >> X86_MODRM_REG_SHIFT) & X86_MODRM_REG_SMASK) | pVCpu->iem.s.uRexReg, u32Cast);
@@ -10673 +10751 @@
             IEM_MC_LOCAL(RTGCPTR, GCPtrEffSrc);
             IEM_MC_CALC_RM_EFF_ADDR(GCPtrEffSrc, bRm, 0);
+            IEMOP_HLP_DONE_DECODING_NO_LOCK_PREFIX();
             IEM_MC_STORE_GREG_U64(((bRm >> X86_MODRM_REG_SHIFT) & X86_MODRM_REG_SMASK) | pVCpu->iem.s.uRexReg, GCPtrEffSrc);
             IEM_MC_ADVANCE_RIP();
@@ -10688 +10767 @@
 
     uint8_t bRm; IEM_OPCODE_GET_NEXT_U8(&bRm);
-    IEMOP_HLP_NO_LOCK_PREFIX(); /** @todo should probably not be raised until we've fetched all the opcode bytes? */
 
     /*
@@ -10711 +10789 @@
     if ((bRm & X86_MODRM_MOD_MASK) == (3 << X86_MODRM_MOD_SHIFT))
     {
+        IEMOP_HLP_DONE_DECODING_NO_LOCK_PREFIX();
         IEM_MC_BEGIN(2, 0);
         IEM_MC_ARG_CONST(uint8_t, iSRegArg, iSegReg, 0);
@@ -10729 +10808 @@
         IEM_MC_LOCAL(RTGCPTR, GCPtrEffDst);
         IEM_MC_CALC_RM_EFF_ADDR(GCPtrEffDst, bRm, 0);
+        IEMOP_HLP_DONE_DECODING_NO_LOCK_PREFIX();
         IEM_MC_FETCH_MEM_U16(u16Value, pVCpu->iem.s.iEffSeg, GCPtrEffDst);
         IEM_MC_CALL_CIMPL_2(iemCImpl_load_SReg, iSRegArg, u16Value);
@@ -10751 +10831 @@
      *        now until tests show it's checked.. */
     IEMOP_MNEMONIC("pop Ev");
-    IEMOP_HLP_NO_LOCK_PREFIX(); /** @todo should probably not be raised until we've fetched all the opcode bytes? */
 
     /* Register access is relatively easy and can share code. */
@@ -10791 +10870 @@
     rcStrict = iemOpHlpCalcRmEffAddr(pVCpu, bRm, 0, &GCPtrEff);
     Assert(rcStrict == VINF_SUCCESS);
+    IEMOP_HLP_DONE_DECODING_NO_LOCK_PREFIX();
     pCtx->rsp = RspSaved;
 
@@ -10859 +10939 @@
 FNIEMOP_DEF_1(iemOpCommonXchgGRegRax, uint8_t, iReg)
 {
-    IEMOP_HLP_NO_LOCK_PREFIX();
+    IEMOP_HLP_DONE_DECODING_NO_LOCK_PREFIX();
 
     iReg |= pVCpu->iem.s.uRexB;
@@ -10985 +11065 @@
 FNIEMOP_DEF(iemOp_cbw)
 {
-    IEMOP_HLP_NO_LOCK_PREFIX();
+    IEMOP_HLP_DONE_DECODING_NO_LOCK_PREFIX();
     switch (pVCpu->iem.s.enmEffOpSize)
     {
@@ -11032 +11112 @@
 FNIEMOP_DEF(iemOp_cwd)
 {
-    IEMOP_HLP_NO_LOCK_PREFIX();
+    IEMOP_HLP_DONE_DECODING_NO_LOCK_PREFIX();
     switch (pVCpu->iem.s.enmEffOpSize)
     {
@@ -11098 +11178 @@
 {
     IEMOP_MNEMONIC("wait");
-    IEMOP_HLP_NO_LOCK_PREFIX();
+    IEMOP_HLP_DONE_DECODING_NO_LOCK_PREFIX();
 
     IEM_MC_BEGIN(0, 0);
@@ -11112 +11192 @@
 FNIEMOP_DEF(iemOp_pushf_Fv)
 {
-    IEMOP_HLP_NO_LOCK_PREFIX();
+    IEMOP_HLP_DONE_DECODING_NO_LOCK_PREFIX();
     IEMOP_HLP_DEFAULT_64BIT_OP_SIZE();
     return IEM_MC_DEFER_TO_CIMPL_1(iemCImpl_pushf, pVCpu->iem.s.enmEffOpSize);
@@ -11121 +11201 @@
 FNIEMOP_DEF(iemOp_popf_Fv)
 {
-    IEMOP_HLP_NO_LOCK_PREFIX();
+    IEMOP_HLP_DONE_DECODING_NO_LOCK_PREFIX();
     IEMOP_HLP_DEFAULT_64BIT_OP_SIZE();
     return IEM_MC_DEFER_TO_CIMPL_1(iemCImpl_popf, pVCpu->iem.s.enmEffOpSize);
@@ -11131 +11211 @@
 {
     IEMOP_MNEMONIC("sahf");
-    IEMOP_HLP_NO_LOCK_PREFIX();
+    IEMOP_HLP_DONE_DECODING_NO_LOCK_PREFIX();
     if (   pVCpu->iem.s.enmCpuMode == IEMMODE_64BIT
         && !IEM_GET_GUEST_CPU_FEATURES(pVCpu)->fLahfSahf)
@@ -11155 +11235 @@
 {
     IEMOP_MNEMONIC("lahf");
-    IEMOP_HLP_NO_LOCK_PREFIX();
+    IEMOP_HLP_DONE_DECODING_NO_LOCK_PREFIX();
     if (   pVCpu->iem.s.enmCpuMode == IEMMODE_64BIT
         && !IEM_GET_GUEST_CPU_FEATURES(pVCpu)->fLahfSahf)
@@ -11191 +11271 @@
             IEM_NOT_REACHED_DEFAULT_CASE_RET(); \
         } \
-        IEMOP_HLP_NO_LOCK_PREFIX(); \
+        IEMOP_HLP_DONE_DECODING_NO_LOCK_PREFIX(); \
     } while (0)
 
@@ -11352 +11432 @@
 FNIEMOP_DEF(iemOp_movsb_Xb_Yb)
 {
-    IEMOP_HLP_NO_LOCK_PREFIX();
+    IEMOP_HLP_DONE_DECODING_NO_LOCK_PREFIX();
 
     /*
@@ -11387 +11467 @@
 FNIEMOP_DEF(iemOp_movswd_Xv_Yv)
 {
-    IEMOP_HLP_NO_LOCK_PREFIX();
+    IEMOP_HLP_DONE_DECODING_NO_LOCK_PREFIX();
 
     /*
@@ -11499 +11579 @@
 FNIEMOP_DEF(iemOp_cmpsb_Xb_Yb)
 {
-    IEMOP_HLP_NO_LOCK_PREFIX();
+    IEMOP_HLP_DONE_DECODING_NO_LOCK_PREFIX();
 
     /*
@@ -11546 +11626 @@
 FNIEMOP_DEF(iemOp_cmpswd_Xv_Yv)
 {
-    IEMOP_HLP_NO_LOCK_PREFIX();
+    IEMOP_HLP_DONE_DECODING_NO_LOCK_PREFIX();
 
     /*
@@ -11701 +11781 @@
 FNIEMOP_DEF(iemOp_stosb_Yb_AL)
 {
-    IEMOP_HLP_NO_LOCK_PREFIX();
+    IEMOP_HLP_DONE_DECODING_NO_LOCK_PREFIX();
 
     /*
@@ -11736 +11816 @@
 FNIEMOP_DEF(iemOp_stoswd_Yv_eAX)
 {
-    IEMOP_HLP_NO_LOCK_PREFIX();
+    IEMOP_HLP_DONE_DECODING_NO_LOCK_PREFIX();
 
     /*
@@ -11837 +11917 @@
 FNIEMOP_DEF(iemOp_lodsb_AL_Xb)
 {
-    IEMOP_HLP_NO_LOCK_PREFIX();
+    IEMOP_HLP_DONE_DECODING_NO_LOCK_PREFIX();
 
     /*
@@ -11872 +11952 @@
 FNIEMOP_DEF(iemOp_lodswd_eAX_Xv)
 {
-    IEMOP_HLP_NO_LOCK_PREFIX();
+    IEMOP_HLP_DONE_DECODING_NO_LOCK_PREFIX();
 
     /*
@@ -11979 +12059 @@
 FNIEMOP_DEF(iemOp_scasb_AL_Xb)
 {
-    IEMOP_HLP_NO_LOCK_PREFIX();
+    IEMOP_HLP_DONE_DECODING_NO_LOCK_PREFIX();
 
     /*
@@ -12025 +12105 @@
 FNIEMOP_DEF(iemOp_scaswd_eAX_Xv)
 {
-    IEMOP_HLP_NO_LOCK_PREFIX();
+    IEMOP_HLP_DONE_DECODING_NO_LOCK_PREFIX();
 
     /*
@@ -12146 +12226 @@
 {
     uint8_t u8Imm; IEM_OPCODE_GET_NEXT_U8(&u8Imm);
-    IEMOP_HLP_NO_LOCK_PREFIX();
+    IEMOP_HLP_DONE_DECODING_NO_LOCK_PREFIX();
 
     IEM_MC_BEGIN(0, 1);
@@ -12232 +12312 @@
         {
             uint16_t u16Imm; IEM_OPCODE_GET_NEXT_U16(&u16Imm);
-            IEMOP_HLP_NO_LOCK_PREFIX();
+            IEMOP_HLP_DONE_DECODING_NO_LOCK_PREFIX();
 
             IEM_MC_BEGIN(0, 1);
@@ -12245 +12325 @@
         {
             uint32_t u32Imm; IEM_OPCODE_GET_NEXT_U32(&u32Imm);
-            IEMOP_HLP_NO_LOCK_PREFIX();
+            IEMOP_HLP_DONE_DECODING_NO_LOCK_PREFIX();
 
             IEM_MC_BEGIN(0, 1);
@@ -12257 +12337 @@
         {
             uint64_t u64Imm; IEM_OPCODE_GET_NEXT_U64(&u64Imm); /* 64-bit immediate! */
-            IEMOP_HLP_NO_LOCK_PREFIX();
+            IEMOP_HLP_DONE_DECODING_NO_LOCK_PREFIX();
 
             IEM_MC_BEGIN(0, 1);
@@ -12360 +12440 @@
         /* register */
         uint8_t cShift; IEM_OPCODE_GET_NEXT_U8(&cShift);
-        IEMOP_HLP_NO_LOCK_PREFIX();
+        IEMOP_HLP_DONE_DECODING_NO_LOCK_PREFIX();
         IEM_MC_BEGIN(3, 0);
         IEM_MC_ARG(uint8_t *,       pu8Dst,            0);
@@ -12374 +12454 @@
     {
         /* memory */
-        IEMOP_HLP_NO_LOCK_PREFIX(); /** @todo should probably not be raised until we've fetched all the opcode bytes? */
         IEM_MC_BEGIN(3, 2);
         IEM_MC_ARG(uint8_t *,   pu8Dst,    0);
@@ -12384 +12463 @@
         uint8_t cShift; IEM_OPCODE_GET_NEXT_U8(&cShift);
         IEM_MC_ASSIGN(cShiftArg, cShift);
+        IEMOP_HLP_DONE_DECODING_NO_LOCK_PREFIX();
         IEM_MC_MEM_MAP(pu8Dst, IEM_ACCESS_DATA_RW, pVCpu->iem.s.iEffSeg, GCPtrEffDst, 0 /*arg*/);
         IEM_MC_FETCH_EFLAGS(EFlags);
@@ -12421 +12501 @@
         /* register */
         uint8_t cShift; IEM_OPCODE_GET_NEXT_U8(&cShift);
-        IEMOP_HLP_NO_LOCK_PREFIX();
+        IEMOP_HLP_DONE_DECODING_NO_LOCK_PREFIX();
         switch (pVCpu->iem.s.enmEffOpSize)
         {
@@ -12467 +12547 @@
     {
         /* memory */
-        IEMOP_HLP_NO_LOCK_PREFIX(); /** @todo should probably not be raised until we've fetched all the opcode bytes? */
         switch (pVCpu->iem.s.enmEffOpSize)
         {
@@ -12480 +12559 @@
                 uint8_t cShift; IEM_OPCODE_GET_NEXT_U8(&cShift);
                 IEM_MC_ASSIGN(cShiftArg, cShift);
+                IEMOP_HLP_DONE_DECODING_NO_LOCK_PREFIX();
                 IEM_MC_MEM_MAP(pu16Dst, IEM_ACCESS_DATA_RW, pVCpu->iem.s.iEffSeg, GCPtrEffDst, 0 /*arg*/);
                 IEM_MC_FETCH_EFLAGS(EFlags);
@@ -12500 +12580 @@
                 uint8_t cShift; IEM_OPCODE_GET_NEXT_U8(&cShift);
                 IEM_MC_ASSIGN(cShiftArg, cShift);
+                IEMOP_HLP_DONE_DECODING_NO_LOCK_PREFIX();
                 IEM_MC_MEM_MAP(pu32Dst, IEM_ACCESS_DATA_RW, pVCpu->iem.s.iEffSeg, GCPtrEffDst, 0 /*arg*/);
                 IEM_MC_FETCH_EFLAGS(EFlags);
@@ -12520 +12601 @@
                 uint8_t cShift; IEM_OPCODE_GET_NEXT_U8(&cShift);
                 IEM_MC_ASSIGN(cShiftArg, cShift);
+                IEMOP_HLP_DONE_DECODING_NO_LOCK_PREFIX();
                 IEM_MC_MEM_MAP(pu64Dst, IEM_ACCESS_DATA_RW, pVCpu->iem.s.iEffSeg, GCPtrEffDst, 0 /*arg*/);
                 IEM_MC_FETCH_EFLAGS(EFlags);
@@ -12541 +12623 @@
     IEMOP_MNEMONIC("retn Iw");
     uint16_t u16Imm; IEM_OPCODE_GET_NEXT_U16(&u16Imm);
-    IEMOP_HLP_NO_LOCK_PREFIX();
+    IEMOP_HLP_DONE_DECODING_NO_LOCK_PREFIX();
     IEMOP_HLP_DEFAULT_64BIT_OP_SIZE();
     return IEM_MC_DEFER_TO_CIMPL_2(iemCImpl_retn, pVCpu->iem.s.enmEffOpSize, u16Imm);
@@ -12552 +12634 @@
     IEMOP_MNEMONIC("retn");
     IEMOP_HLP_DEFAULT_64BIT_OP_SIZE();
-    IEMOP_HLP_NO_LOCK_PREFIX();
+    IEMOP_HLP_DONE_DECODING_NO_LOCK_PREFIX();
     return IEM_MC_DEFER_TO_CIMPL_2(iemCImpl_retn, pVCpu->iem.s.enmEffOpSize, 0);
 }
@@ -12622 +12704 @@
 {
     uint8_t bRm; IEM_OPCODE_GET_NEXT_U8(&bRm);
-    IEMOP_HLP_NO_LOCK_PREFIX(); /** @todo should probably not be raised until we've fetched all the opcode bytes? */
     if ((bRm & X86_MODRM_REG_MASK) != (0 << X86_MODRM_REG_SHIFT)) /* only mov Eb,Ib in this group. */
         return IEMOP_RAISE_INVALID_OPCODE();
@@ -12631 +12712 @@
         /* register access */
         uint8_t u8Imm; IEM_OPCODE_GET_NEXT_U8(&u8Imm);
+        IEMOP_HLP_DONE_DECODING_NO_LOCK_PREFIX();
         IEM_MC_BEGIN(0, 0);
         IEM_MC_STORE_GREG_U8((bRm & X86_MODRM_RM_MASK) | pVCpu->iem.s.uRexB, u8Imm);
@@ -12643 +12725 @@
         IEM_MC_CALC_RM_EFF_ADDR(GCPtrEffDst, bRm, 1);
         uint8_t u8Imm; IEM_OPCODE_GET_NEXT_U8(&u8Imm);
+        IEMOP_HLP_DONE_DECODING_NO_LOCK_PREFIX();
         IEM_MC_STORE_MEM_U8(pVCpu->iem.s.iEffSeg, GCPtrEffDst, u8Imm);
         IEM_MC_ADVANCE_RIP();
@@ -12655 +12738 @@
 {
     uint8_t bRm; IEM_OPCODE_GET_NEXT_U8(&bRm);
-    IEMOP_HLP_NO_LOCK_PREFIX(); /** @todo should probably not be raised until we've fetched all the opcode bytes? */
     if ((bRm & X86_MODRM_REG_MASK) != (0 << X86_MODRM_REG_SHIFT)) /* only mov Eb,Ib in this group. */
         return IEMOP_RAISE_INVALID_OPCODE();
@@ -12668 +12750 @@
                 IEM_MC_BEGIN(0, 0);
                 uint16_t u16Imm; IEM_OPCODE_GET_NEXT_U16(&u16Imm);
+                IEMOP_HLP_DONE_DECODING_NO_LOCK_PREFIX();
                 IEM_MC_STORE_GREG_U16((bRm & X86_MODRM_RM_MASK) | pVCpu->iem.s.uRexB, u16Imm);
                 IEM_MC_ADVANCE_RIP();
@@ -12676 +12759 @@
                 IEM_MC_BEGIN(0, 0);
                 uint32_t u32Imm; IEM_OPCODE_GET_NEXT_U32(&u32Imm);
+                IEMOP_HLP_DONE_DECODING_NO_LOCK_PREFIX();
                 IEM_MC_STORE_GREG_U32((bRm & X86_MODRM_RM_MASK) | pVCpu->iem.s.uRexB, u32Imm);
                 IEM_MC_ADVANCE_RIP();
@@ -12684 +12768 @@
                 IEM_MC_BEGIN(0, 0);
                 uint64_t u64Imm; IEM_OPCODE_GET_NEXT_S32_SX_U64(&u64Imm);
+                IEMOP_HLP_DONE_DECODING_NO_LOCK_PREFIX();
                 IEM_MC_STORE_GREG_U64((bRm & X86_MODRM_RM_MASK) | pVCpu->iem.s.uRexB, u64Imm);
                 IEM_MC_ADVANCE_RIP();
@@ -12702 +12787 @@
                 IEM_MC_CALC_RM_EFF_ADDR(GCPtrEffDst, bRm, 2);
                 uint16_t u16Imm; IEM_OPCODE_GET_NEXT_U16(&u16Imm);
+                IEMOP_HLP_DONE_DECODING_NO_LOCK_PREFIX();
                 IEM_MC_STORE_MEM_U16(pVCpu->iem.s.iEffSeg, GCPtrEffDst, u16Imm);
                 IEM_MC_ADVANCE_RIP();
@@ -12712 +12798 @@
                 IEM_MC_CALC_RM_EFF_ADDR(GCPtrEffDst, bRm, 4);
                 uint32_t u32Imm; IEM_OPCODE_GET_NEXT_U32(&u32Imm);
+                IEMOP_HLP_DONE_DECODING_NO_LOCK_PREFIX();
                 IEM_MC_STORE_MEM_U32(pVCpu->iem.s.iEffSeg, GCPtrEffDst, u32Imm);
                 IEM_MC_ADVANCE_RIP();
@@ -12722 +12809 @@
                 IEM_MC_CALC_RM_EFF_ADDR(GCPtrEffDst, bRm, 4);
                 uint64_t u64Imm; IEM_OPCODE_GET_NEXT_S32_SX_U64(&u64Imm);
+                IEMOP_HLP_DONE_DECODING_NO_LOCK_PREFIX();
                 IEM_MC_STORE_MEM_U64(pVCpu->iem.s.iEffSeg, GCPtrEffDst, u64Imm);
                 IEM_MC_ADVANCE_RIP();
@@ -12741 +12829 @@
     IEMOP_HLP_MIN_186();
     IEMOP_HLP_DEFAULT_64BIT_OP_SIZE();
-    IEMOP_HLP_NO_LOCK_PREFIX();
     uint16_t cbFrame;        IEM_OPCODE_GET_NEXT_U16(&cbFrame);
     uint8_t  u8NestingLevel; IEM_OPCODE_GET_NEXT_U8(&u8NestingLevel);
+    IEMOP_HLP_DONE_DECODING_NO_LOCK_PREFIX();
     return IEM_MC_DEFER_TO_CIMPL_3(iemCImpl_enter, pVCpu->iem.s.enmEffOpSize, cbFrame, u8NestingLevel);
 }
@@ -12754 +12842 @@
     IEMOP_HLP_MIN_186();
     IEMOP_HLP_DEFAULT_64BIT_OP_SIZE();
-    IEMOP_HLP_NO_LOCK_PREFIX();
+    IEMOP_HLP_DONE_DECODING_NO_LOCK_PREFIX();
     return IEM_MC_DEFER_TO_CIMPL_1(iemCImpl_leave, pVCpu->iem.s.enmEffOpSize);
 }
@@ -12764 +12852 @@
     IEMOP_MNEMONIC("retf Iw");
     uint16_t u16Imm; IEM_OPCODE_GET_NEXT_U16(&u16Imm);
-    IEMOP_HLP_NO_LOCK_PREFIX();
+    IEMOP_HLP_DONE_DECODING_NO_LOCK_PREFIX();
     IEMOP_HLP_DEFAULT_64BIT_OP_SIZE();
     return IEM_MC_DEFER_TO_CIMPL_2(iemCImpl_retf, pVCpu->iem.s.enmEffOpSize, u16Imm);
@@ -12774 +12862 @@
 {
     IEMOP_MNEMONIC("retf");
-    IEMOP_HLP_NO_LOCK_PREFIX();
+    IEMOP_HLP_DONE_DECODING_NO_LOCK_PREFIX();
     IEMOP_HLP_DEFAULT_64BIT_OP_SIZE();
     return IEM_MC_DEFER_TO_CIMPL_2(iemCImpl_retf, pVCpu->iem.s.enmEffOpSize, 0);
@@ -12816 +12904 @@
 {
     IEMOP_MNEMONIC("iret");
-    IEMOP_HLP_NO_LOCK_PREFIX();
+    IEMOP_HLP_DONE_DECODING_NO_LOCK_PREFIX();
     return IEM_MC_DEFER_TO_CIMPL_1(iemCImpl_iret, pVCpu->iem.s.enmEffOpSize);
 }
@@ -12843 +12931 @@
     {
         /* register */
-        IEMOP_HLP_NO_LOCK_PREFIX();
+        IEMOP_HLP_DONE_DECODING_NO_LOCK_PREFIX();
         IEM_MC_BEGIN(3, 0);
         IEM_MC_ARG(uint8_t *,       pu8Dst,             0);
@@ -12857 +12945 @@
     {
         /* memory */
-        IEMOP_HLP_NO_LOCK_PREFIX(); /** @todo should probably not be raised until we've fetched all the opcode bytes? */
         IEM_MC_BEGIN(3, 2);
         IEM_MC_ARG(uint8_t *,       pu8Dst,             0);
@@ -12865 +12952 @@
 
         IEM_MC_CALC_RM_EFF_ADDR(GCPtrEffDst, bRm, 0);
+        IEMOP_HLP_DONE_DECODING_NO_LOCK_PREFIX();
         IEM_MC_MEM_MAP(pu8Dst, IEM_ACCESS_DATA_RW, pVCpu->iem.s.iEffSeg, GCPtrEffDst, 0 /*arg*/);
         IEM_MC_FETCH_EFLAGS(EFlags);
@@ -12901 +12989 @@
     {
         /* register */
-        IEMOP_HLP_NO_LOCK_PREFIX();
+        IEMOP_HLP_DONE_DECODING_NO_LOCK_PREFIX();
         switch (pVCpu->iem.s.enmEffOpSize)
         {
@@ -12947 +13035 @@
     {
         /* memory */
-        IEMOP_HLP_NO_LOCK_PREFIX(); /** @todo should probably not be raised until we've fetched all the opcode bytes? */
         switch (pVCpu->iem.s.enmEffOpSize)
         {
@@ -12958 +13045 @@
 
                 IEM_MC_CALC_RM_EFF_ADDR(GCPtrEffDst, bRm, 0);
+                IEMOP_HLP_DONE_DECODING_NO_LOCK_PREFIX();
                 IEM_MC_MEM_MAP(pu16Dst, IEM_ACCESS_DATA_RW, pVCpu->iem.s.iEffSeg, GCPtrEffDst, 0 /*arg*/);
                 IEM_MC_FETCH_EFLAGS(EFlags);
@@ -12976 +13064 @@
 
                 IEM_MC_CALC_RM_EFF_ADDR(GCPtrEffDst, bRm, 0);
+                IEMOP_HLP_DONE_DECODING_NO_LOCK_PREFIX();
                 IEM_MC_MEM_MAP(pu32Dst, IEM_ACCESS_DATA_RW, pVCpu->iem.s.iEffSeg, GCPtrEffDst, 0 /*arg*/);
                 IEM_MC_FETCH_EFLAGS(EFlags);
@@ -12994 +13083 @@
 
                 IEM_MC_CALC_RM_EFF_ADDR(GCPtrEffDst, bRm, 0);
+                IEMOP_HLP_DONE_DECODING_NO_LOCK_PREFIX();
                 IEM_MC_MEM_MAP(pu64Dst, IEM_ACCESS_DATA_RW, pVCpu->iem.s.iEffSeg, GCPtrEffDst, 0 /*arg*/);
                 IEM_MC_FETCH_EFLAGS(EFlags);
@@ -13032 +13122 @@
     {
         /* register */
-        IEMOP_HLP_NO_LOCK_PREFIX();
+        IEMOP_HLP_DONE_DECODING_NO_LOCK_PREFIX();
         IEM_MC_BEGIN(3, 0);
         IEM_MC_ARG(uint8_t *,   pu8Dst,     0);
@@ -13047 +13137 @@
     {
         /* memory */
-        IEMOP_HLP_NO_LOCK_PREFIX(); /** @todo should probably not be raised until we've fetched all the opcode bytes? */
         IEM_MC_BEGIN(3, 2);
         IEM_MC_ARG(uint8_t *,   pu8Dst,          0);
@@ -13055 +13144 @@
 
         IEM_MC_CALC_RM_EFF_ADDR(GCPtrEffDst, bRm, 0);
+        IEMOP_HLP_DONE_DECODING_NO_LOCK_PREFIX();
         IEM_MC_FETCH_GREG_U8(cShiftArg, X86_GREG_xCX);
         IEM_MC_MEM_MAP(pu8Dst, IEM_ACCESS_DATA_RW, pVCpu->iem.s.iEffSeg, GCPtrEffDst, 0 /*arg*/);
@@ -13091 +13181 @@
     {
         /* register */
-        IEMOP_HLP_NO_LOCK_PREFIX();
+        IEMOP_HLP_DONE_DECODING_NO_LOCK_PREFIX();
         switch (pVCpu->iem.s.enmEffOpSize)
         {
@@ -13140 +13230 @@
     {
         /* memory */
-        IEMOP_HLP_NO_LOCK_PREFIX(); /** @todo should probably not be raised until we've fetched all the opcode bytes? */
         switch (pVCpu->iem.s.enmEffOpSize)
         {
@@ -13151 +13240 @@
 
                 IEM_MC_CALC_RM_EFF_ADDR(GCPtrEffDst, bRm, 0);
+                IEMOP_HLP_DONE_DECODING_NO_LOCK_PREFIX();
                 IEM_MC_FETCH_GREG_U8(cShiftArg, X86_GREG_xCX);
                 IEM_MC_MEM_MAP(pu16Dst, IEM_ACCESS_DATA_RW, pVCpu->iem.s.iEffSeg, GCPtrEffDst, 0 /*arg*/);
@@ -13170 +13260 @@
 
                 IEM_MC_CALC_RM_EFF_ADDR(GCPtrEffDst, bRm, 0);
+                IEMOP_HLP_DONE_DECODING_NO_LOCK_PREFIX();
                 IEM_MC_FETCH_GREG_U8(cShiftArg, X86_GREG_xCX);
                 IEM_MC_MEM_MAP(pu32Dst, IEM_ACCESS_DATA_RW, pVCpu->iem.s.iEffSeg, GCPtrEffDst, 0 /*arg*/);
@@ -13189 +13280 @@
 
                 IEM_MC_CALC_RM_EFF_ADDR(GCPtrEffDst, bRm, 0);
+                IEMOP_HLP_DONE_DECODING_NO_LOCK_PREFIX();
                 IEM_MC_FETCH_GREG_U8(cShiftArg, X86_GREG_xCX);
                 IEM_MC_MEM_MAP(pu64Dst, IEM_ACCESS_DATA_RW, pVCpu->iem.s.iEffSeg, GCPtrEffDst, 0 /*arg*/);
@@ -13210 +13302 @@
     IEMOP_MNEMONIC("aam Ib");
     uint8_t bImm; IEM_OPCODE_GET_NEXT_U8(&bImm);
-    IEMOP_HLP_NO_LOCK_PREFIX();
+    IEMOP_HLP_DONE_DECODING_NO_LOCK_PREFIX();
     IEMOP_HLP_NO_64BIT();
     if (!bImm)
@@ -13223 +13315 @@
     IEMOP_MNEMONIC("aad Ib");
     uint8_t bImm; IEM_OPCODE_GET_NEXT_U8(&bImm);
-    IEMOP_HLP_NO_LOCK_PREFIX();
+    IEMOP_HLP_DONE_DECODING_NO_LOCK_PREFIX();
     IEMOP_HLP_NO_64BIT();
     return IEM_MC_DEFER_TO_CIMPL_1(iemCImpl_aad, bImm);
@@ -13254 +13346 @@
 {
     IEMOP_MNEMONIC("xlat");
-    IEMOP_HLP_NO_LOCK_PREFIX();
+    IEMOP_HLP_DONE_DECODING_NO_LOCK_PREFIX();
     switch (pVCpu->iem.s.enmEffAddrMode)
     {
@@ -16406 +16498 @@
     IEMOP_MNEMONIC("loopne Jb");
     int8_t i8Imm; IEM_OPCODE_GET_NEXT_S8(&i8Imm);
-    IEMOP_HLP_NO_LOCK_PREFIX();
+    IEMOP_HLP_DONE_DECODING_NO_LOCK_PREFIX();
     IEMOP_HLP_DEFAULT_64BIT_OP_SIZE();
 
@@ -16454 +16546 @@
     IEMOP_MNEMONIC("loope Jb");
     int8_t i8Imm; IEM_OPCODE_GET_NEXT_S8(&i8Imm);
-    IEMOP_HLP_NO_LOCK_PREFIX();
+    IEMOP_HLP_DONE_DECODING_NO_LOCK_PREFIX();
     IEMOP_HLP_DEFAULT_64BIT_OP_SIZE();
 
@@ -16502 +16594 @@
     IEMOP_MNEMONIC("loop Jb");
     int8_t i8Imm; IEM_OPCODE_GET_NEXT_S8(&i8Imm);
-    IEMOP_HLP_NO_LOCK_PREFIX();
+    IEMOP_HLP_DONE_DECODING_NO_LOCK_PREFIX();
     IEMOP_HLP_DEFAULT_64BIT_OP_SIZE();
 
@@ -16577 +16669 @@
     IEMOP_MNEMONIC("jecxz Jb");
     int8_t i8Imm; IEM_OPCODE_GET_NEXT_S8(&i8Imm);
-    IEMOP_HLP_NO_LOCK_PREFIX();
+    IEMOP_HLP_DONE_DECODING_NO_LOCK_PREFIX();
     IEMOP_HLP_DEFAULT_64BIT_OP_SIZE();
 
@@ -16622 +16714 @@
     IEMOP_MNEMONIC("in eAX,Ib");
     uint8_t u8Imm; IEM_OPCODE_GET_NEXT_U8(&u8Imm);
-    IEMOP_HLP_NO_LOCK_PREFIX();
+    IEMOP_HLP_DONE_DECODING_NO_LOCK_PREFIX();
     return IEM_MC_DEFER_TO_CIMPL_2(iemCImpl_in, u8Imm, 1);
 }
@@ -16632 +16724 @@
     IEMOP_MNEMONIC("in eAX,Ib");
     uint8_t u8Imm; IEM_OPCODE_GET_NEXT_U8(&u8Imm);
-    IEMOP_HLP_NO_LOCK_PREFIX();
+    IEMOP_HLP_DONE_DECODING_NO_LOCK_PREFIX();
     return IEM_MC_DEFER_TO_CIMPL_2(iemCImpl_in, u8Imm, pVCpu->iem.s.enmEffOpSize == IEMMODE_16BIT ? 2 : 4);
 }
@@ -16642 +16734 @@
     IEMOP_MNEMONIC("out Ib,AL");
     uint8_t u8Imm; IEM_OPCODE_GET_NEXT_U8(&u8Imm);
-    IEMOP_HLP_NO_LOCK_PREFIX();
+    IEMOP_HLP_DONE_DECODING_NO_LOCK_PREFIX();
     return IEM_MC_DEFER_TO_CIMPL_2(iemCImpl_out, u8Imm, 1);
 }
@@ -16652 +16744 @@
     IEMOP_MNEMONIC("out Ib,eAX");
     uint8_t u8Imm; IEM_OPCODE_GET_NEXT_U8(&u8Imm);
-    IEMOP_HLP_NO_LOCK_PREFIX();
+    IEMOP_HLP_DONE_DECODING_NO_LOCK_PREFIX();
     return IEM_MC_DEFER_TO_CIMPL_2(iemCImpl_out, u8Imm, pVCpu->iem.s.enmEffOpSize == IEMMODE_16BIT ? 2 : 4);
 }
@@ -16731 +16823 @@
         IEM_OPCODE_GET_NEXT_U16_ZX_U32(&offSeg);
     uint16_t uSel;  IEM_OPCODE_GET_NEXT_U16(&uSel);
-    IEMOP_HLP_NO_LOCK_PREFIX();
+    IEMOP_HLP_DONE_DECODING_NO_LOCK_PREFIX();
     return IEM_MC_DEFER_TO_CIMPL_3(iemCImpl_FarJmp, uSel, offSeg, pVCpu->iem.s.enmEffOpSize);
 }
@@ -16741 +16833 @@
     IEMOP_MNEMONIC("jmp Jb");
     int8_t i8Imm; IEM_OPCODE_GET_NEXT_S8(&i8Imm);
-    IEMOP_HLP_NO_LOCK_PREFIX();
+    IEMOP_HLP_DONE_DECODING_NO_LOCK_PREFIX();
     IEMOP_HLP_DEFAULT_64BIT_OP_SIZE();
 
@@ -16755 +16847 @@
 {
     IEMOP_MNEMONIC("in  AL,DX");
-    IEMOP_HLP_NO_LOCK_PREFIX();
+    IEMOP_HLP_DONE_DECODING_NO_LOCK_PREFIX();
     return IEM_MC_DEFER_TO_CIMPL_1(iemCImpl_in_eAX_DX, 1);
 }
@@ -16764 +16856 @@
 {
     IEMOP_MNEMONIC("in  eAX,DX");
-    IEMOP_HLP_NO_LOCK_PREFIX();
+    IEMOP_HLP_DONE_DECODING_NO_LOCK_PREFIX();
     return IEM_MC_DEFER_TO_CIMPL_1(iemCImpl_in_eAX_DX, pVCpu->iem.s.enmEffOpSize == IEMMODE_16BIT ? 2 : 4);
 }
@@ -16773 +16865 @@
 {
     IEMOP_MNEMONIC("out DX,AL");
-    IEMOP_HLP_NO_LOCK_PREFIX();
+    IEMOP_HLP_DONE_DECODING_NO_LOCK_PREFIX();
     return IEM_MC_DEFER_TO_CIMPL_1(iemCImpl_out_DX_eAX, 1);
 }
@@ -16782 +16874 @@
 {
     IEMOP_MNEMONIC("out DX,eAX");
-    IEMOP_HLP_NO_LOCK_PREFIX();
+    IEMOP_HLP_DONE_DECODING_NO_LOCK_PREFIX();
     return IEM_MC_DEFER_TO_CIMPL_1(iemCImpl_out_DX_eAX, pVCpu->iem.s.enmEffOpSize == IEMMODE_16BIT ? 2 : 4);
 }
@@ -16837 +16929 @@
 FNIEMOP_DEF(iemOp_hlt)
 {
-    IEMOP_HLP_NO_LOCK_PREFIX();
+    IEMOP_HLP_DONE_DECODING_NO_LOCK_PREFIX();
     return IEM_MC_DEFER_TO_CIMPL_0(iemCImpl_hlt);
 }
@@ -16846 +16938 @@
 {
     IEMOP_MNEMONIC("cmc");
-    IEMOP_HLP_NO_LOCK_PREFIX();
+    IEMOP_HLP_DONE_DECODING_NO_LOCK_PREFIX();
     IEM_MC_BEGIN(0, 0);
     IEM_MC_FLIP_EFL_BIT(X86_EFL_CF);
@@ -16990 +17082 @@
         /* register access */
         uint8_t u8Imm; IEM_OPCODE_GET_NEXT_U8(&u8Imm);
-        IEMOP_HLP_NO_LOCK_PREFIX();
+        IEMOP_HLP_DONE_DECODING_NO_LOCK_PREFIX();
 
         IEM_MC_BEGIN(3, 0);
@@ -17005 +17097 @@
     {
         /* memory access. */
-        IEMOP_HLP_NO_LOCK_PREFIX(); /** @todo should probably not be raised until we've fetched all the opcode bytes? */
-
         IEM_MC_BEGIN(3, 2);
         IEM_MC_ARG(uint8_t *,       pu8Dst,             0);
@@ -17016 +17106 @@
         uint8_t u8Imm; IEM_OPCODE_GET_NEXT_U8(&u8Imm);
         IEM_MC_ASSIGN(u8Src, u8Imm);
+        IEMOP_HLP_DONE_DECODING_NO_LOCK_PREFIX();
         IEM_MC_MEM_MAP(pu8Dst, IEM_ACCESS_DATA_R, pVCpu->iem.s.iEffSeg, GCPtrEffDst, 0 /*arg*/);
         IEM_MC_FETCH_EFLAGS(EFlags);
@@ -17033 +17124 @@
 {
     IEMOP_MNEMONIC("test Ev,Iv");
-    IEMOP_HLP_NO_LOCK_PREFIX(); /** @todo should probably not be raised until we've fetched all the opcode bytes? */
     IEMOP_VERIFICATION_UNDEFINED_EFLAGS(X86_EFL_AF);
 
@@ -17039 +17129 @@
     {
         /* register access */
+        IEMOP_HLP_DONE_DECODING_NO_LOCK_PREFIX();
         switch (pVCpu->iem.s.enmEffOpSize)
         {
@@ -17106 +17197 @@
                 uint16_t u16Imm; IEM_OPCODE_GET_NEXT_U16(&u16Imm);
                 IEM_MC_ASSIGN(u16Src, u16Imm);
+                IEMOP_HLP_DONE_DECODING_NO_LOCK_PREFIX();
                 IEM_MC_MEM_MAP(pu16Dst, IEM_ACCESS_DATA_R, pVCpu->iem.s.iEffSeg, GCPtrEffDst, 0 /*arg*/);
                 IEM_MC_FETCH_EFLAGS(EFlags);
@@ -17128 +17220 @@
                 uint32_t u32Imm; IEM_OPCODE_GET_NEXT_U32(&u32Imm);
                 IEM_MC_ASSIGN(u32Src, u32Imm);
+                IEMOP_HLP_DONE_DECODING_NO_LOCK_PREFIX();
                 IEM_MC_MEM_MAP(pu32Dst, IEM_ACCESS_DATA_R, pVCpu->iem.s.iEffSeg, GCPtrEffDst, 0 /*arg*/);
                 IEM_MC_FETCH_EFLAGS(EFlags);
@@ -17150 +17243 @@
                 uint64_t u64Imm; IEM_OPCODE_GET_NEXT_S32_SX_U64(&u64Imm);
                 IEM_MC_ASSIGN(u64Src, u64Imm);
+                IEMOP_HLP_DONE_DECODING_NO_LOCK_PREFIX();
                 IEM_MC_MEM_MAP(pu64Dst, IEM_ACCESS_DATA_R, pVCpu->iem.s.iEffSeg, GCPtrEffDst, 0 /*arg*/);
                 IEM_MC_FETCH_EFLAGS(EFlags);
@@ -17170 +17264 @@
 FNIEMOP_DEF_2(iemOpCommonGrp3MulDivEb, uint8_t, bRm, PFNIEMAIMPLMULDIVU8, pfnU8)
 {
-    IEMOP_HLP_NO_LOCK_PREFIX(); /** @todo should probably not be raised until we've fetched all the opcode bytes? */
-
     if ((bRm & X86_MODRM_MOD_MASK) == (3 << X86_MODRM_MOD_SHIFT))
     {
         /* register access */
-        IEMOP_HLP_NO_LOCK_PREFIX();
+        IEMOP_HLP_DONE_DECODING_NO_LOCK_PREFIX();
         IEM_MC_BEGIN(3, 1);
         IEM_MC_ARG(uint16_t *,      pu16AX,     0);
@@ -17197 +17289 @@
     {
         /* memory access. */
-        IEMOP_HLP_NO_LOCK_PREFIX(); /** @todo should probably not be raised until we've fetched all the opcode bytes? */
-
         IEM_MC_BEGIN(3, 2);
         IEM_MC_ARG(uint16_t *,      pu16AX,     0);
@@ -17207 +17297 @@
 
         IEM_MC_CALC_RM_EFF_ADDR(GCPtrEffDst, bRm, 0);
+        IEMOP_HLP_DONE_DECODING_NO_LOCK_PREFIX();
         IEM_MC_FETCH_MEM_U8(u8Value, pVCpu->iem.s.iEffSeg, GCPtrEffDst);
         IEM_MC_REF_GREG_U16(pu16AX, X86_GREG_xAX);
@@ -17226 +17317 @@
 FNIEMOP_DEF_2(iemOpCommonGrp3MulDivEv, uint8_t, bRm, PCIEMOPMULDIVSIZES, pImpl)
 {
-    IEMOP_HLP_NO_LOCK_PREFIX(); /** @todo should probably not be raised until we've fetched all the opcode bytes? */
     IEMOP_VERIFICATION_UNDEFINED_EFLAGS(X86_EFL_SF | X86_EFL_ZF | X86_EFL_AF | X86_EFL_PF);
 
@@ -17232 +17322 @@
     {
         /* register access */
+        IEMOP_HLP_DONE_DECODING_NO_LOCK_PREFIX();
         switch (pVCpu->iem.s.enmEffOpSize)
         {
             case IEMMODE_16BIT:
             {
-                IEMOP_HLP_NO_LOCK_PREFIX();
+                IEMOP_HLP_DONE_DECODING_NO_LOCK_PREFIX();
                 IEM_MC_BEGIN(4, 1);
                 IEM_MC_ARG(uint16_t *,      pu16AX,     0);
@@ -17261 +17352 @@
             case IEMMODE_32BIT:
             {
-                IEMOP_HLP_NO_LOCK_PREFIX();
+                IEMOP_HLP_DONE_DECODING_NO_LOCK_PREFIX();
                 IEM_MC_BEGIN(4, 1);
                 IEM_MC_ARG(uint32_t *,      pu32AX,     0);
@@ -17288 +17379 @@
             case IEMMODE_64BIT:
             {
-                IEMOP_HLP_NO_LOCK_PREFIX();
+                IEMOP_HLP_DONE_DECODING_NO_LOCK_PREFIX();
                 IEM_MC_BEGIN(4, 1);
                 IEM_MC_ARG(uint64_t *,      pu64AX,     0);
@@ -17321 +17412 @@
             case IEMMODE_16BIT:
             {
-                IEMOP_HLP_NO_LOCK_PREFIX();
                 IEM_MC_BEGIN(4, 2);
                 IEM_MC_ARG(uint16_t *,      pu16AX,     0);
@@ -17331 +17421 @@
 
                 IEM_MC_CALC_RM_EFF_ADDR(GCPtrEffDst, bRm, 0);
+                IEMOP_HLP_DONE_DECODING_NO_LOCK_PREFIX();
                 IEM_MC_FETCH_MEM_U16(u16Value, pVCpu->iem.s.iEffSeg, GCPtrEffDst);
                 IEM_MC_REF_GREG_U16(pu16AX, X86_GREG_xAX);
@@ -17348 +17439 @@
             case IEMMODE_32BIT:
             {
-                IEMOP_HLP_NO_LOCK_PREFIX();
                 IEM_MC_BEGIN(4, 2);
                 IEM_MC_ARG(uint32_t *,      pu32AX,     0);
@@ -17358 +17448 @@
 
                 IEM_MC_CALC_RM_EFF_ADDR(GCPtrEffDst, bRm, 0);
+                IEMOP_HLP_DONE_DECODING_NO_LOCK_PREFIX();
                 IEM_MC_FETCH_MEM_U32(u32Value, pVCpu->iem.s.iEffSeg, GCPtrEffDst);
                 IEM_MC_REF_GREG_U32(pu32AX, X86_GREG_xAX);
@@ -17377 +17468 @@
             case IEMMODE_64BIT:
             {
-                IEMOP_HLP_NO_LOCK_PREFIX();
                 IEM_MC_BEGIN(4, 2);
                 IEM_MC_ARG(uint64_t *,      pu64AX,     0);
@@ -17387 +17477 @@
 
                 IEM_MC_CALC_RM_EFF_ADDR(GCPtrEffDst, bRm, 0);
+                IEMOP_HLP_DONE_DECODING_NO_LOCK_PREFIX();
                 IEM_MC_FETCH_MEM_U64(u64Value, pVCpu->iem.s.iEffSeg, GCPtrEffDst);
                 IEM_MC_REF_GREG_U64(pu64AX, X86_GREG_xAX);
@@ -17487 +17578 @@
 {
     IEMOP_MNEMONIC("clc");
-    IEMOP_HLP_NO_LOCK_PREFIX();
+    IEMOP_HLP_DONE_DECODING_NO_LOCK_PREFIX();
     IEM_MC_BEGIN(0, 0);
     IEM_MC_CLEAR_EFL_BIT(X86_EFL_CF);
@@ -17500 +17591 @@
 {
     IEMOP_MNEMONIC("stc");
-    IEMOP_HLP_NO_LOCK_PREFIX();
+    IEMOP_HLP_DONE_DECODING_NO_LOCK_PREFIX();
     IEM_MC_BEGIN(0, 0);
     IEM_MC_SET_EFL_BIT(X86_EFL_CF);
@@ -17513 +17604 @@
 {
     IEMOP_MNEMONIC("cli");
-    IEMOP_HLP_NO_LOCK_PREFIX();
+    IEMOP_HLP_DONE_DECODING_NO_LOCK_PREFIX();
     return IEM_MC_DEFER_TO_CIMPL_0(iemCImpl_cli);
 }
@@ -17521 +17612 @@
 {
     IEMOP_MNEMONIC("sti");
-    IEMOP_HLP_NO_LOCK_PREFIX();
+    IEMOP_HLP_DONE_DECODING_NO_LOCK_PREFIX();
     return IEM_MC_DEFER_TO_CIMPL_0(iemCImpl_sti);
 }
@@ -17530 +17621 @@
 {
     IEMOP_MNEMONIC("cld");
-    IEMOP_HLP_NO_LOCK_PREFIX();
+    IEMOP_HLP_DONE_DECODING_NO_LOCK_PREFIX();
     IEM_MC_BEGIN(0, 0);
     IEM_MC_CLEAR_EFL_BIT(X86_EFL_DF);
@@ -17543 +17634 @@
 {
     IEMOP_MNEMONIC("std");
-    IEMOP_HLP_NO_LOCK_PREFIX();
+    IEMOP_HLP_DONE_DECODING_NO_LOCK_PREFIX();
     IEM_MC_BEGIN(0, 0);
     IEM_MC_SET_EFL_BIT(X86_EFL_DF);
@@ -17578 +17669 @@
 {
     IEMOP_MNEMONIC("calln Ev");
-    IEMOP_HLP_NO_LOCK_PREFIX(); /** @todo Too early? */
     IEMOP_HLP_DEFAULT_64BIT_OP_SIZE();
 
@@ -17584 +17674 @@
     {
         /* The new RIP is taken from a register. */
+        IEMOP_HLP_DONE_DECODING_NO_LOCK_PREFIX();
         switch (pVCpu->iem.s.enmEffOpSize)
         {
@@ -17623 +17714 @@
                 IEM_MC_LOCAL(RTGCPTR, GCPtrEffSrc);
                 IEM_MC_CALC_RM_EFF_ADDR(GCPtrEffSrc, bRm, 0);
+                IEMOP_HLP_DONE_DECODING_NO_LOCK_PREFIX();
                 IEM_MC_FETCH_MEM_U16(u16Target, pVCpu->iem.s.iEffSeg, GCPtrEffSrc);
                 IEM_MC_CALL_CIMPL_1(iemCImpl_call_16, u16Target);
@@ -17633 +17725 @@
                 IEM_MC_LOCAL(RTGCPTR, GCPtrEffSrc);
                 IEM_MC_CALC_RM_EFF_ADDR(GCPtrEffSrc, bRm, 0);
+                IEMOP_HLP_DONE_DECODING_NO_LOCK_PREFIX();
                 IEM_MC_FETCH_MEM_U32(u32Target, pVCpu->iem.s.iEffSeg, GCPtrEffSrc);
                 IEM_MC_CALL_CIMPL_1(iemCImpl_call_32, u32Target);
@@ -17643 +17736 @@
                 IEM_MC_LOCAL(RTGCPTR, GCPtrEffSrc);
                 IEM_MC_CALC_RM_EFF_ADDR(GCPtrEffSrc, bRm, 0);
+                IEMOP_HLP_DONE_DECODING_NO_LOCK_PREFIX();
                 IEM_MC_FETCH_MEM_U64(u64Target, pVCpu->iem.s.iEffSeg, GCPtrEffSrc);
                 IEM_MC_CALL_CIMPL_1(iemCImpl_call_64, u64Target);
@@ -17738 +17832 @@
 {
     IEMOP_MNEMONIC("jmpn Ev");
-    IEMOP_HLP_NO_LOCK_PREFIX(); /** @todo Too early? */
     IEMOP_HLP_DEFAULT_64BIT_OP_SIZE();
 
@@ -17744 +17837 @@
     {
         /* The new RIP is taken from a register. */
+        IEMOP_HLP_DONE_DECODING_NO_LOCK_PREFIX();
         switch (pVCpu->iem.s.enmEffOpSize)
         {
@@ -17783 +17877 @@
                 IEM_MC_LOCAL(RTGCPTR, GCPtrEffSrc);
                 IEM_MC_CALC_RM_EFF_ADDR(GCPtrEffSrc, bRm, 0);
+                IEMOP_HLP_DONE_DECODING_NO_LOCK_PREFIX();
                 IEM_MC_FETCH_MEM_U16(u16Target, pVCpu->iem.s.iEffSeg, GCPtrEffSrc);
                 IEM_MC_SET_RIP_U16(u16Target);
@@ -17793 +17888 @@
                 IEM_MC_LOCAL(RTGCPTR, GCPtrEffSrc);
                 IEM_MC_CALC_RM_EFF_ADDR(GCPtrEffSrc, bRm, 0);
+                IEMOP_HLP_DONE_DECODING_NO_LOCK_PREFIX();
                 IEM_MC_FETCH_MEM_U32(u32Target, pVCpu->iem.s.iEffSeg, GCPtrEffSrc);
                 IEM_MC_SET_RIP_U32(u32Target);
@@ -17803 +17899 @@
                 IEM_MC_LOCAL(RTGCPTR, GCPtrEffSrc);
                 IEM_MC_CALC_RM_EFF_ADDR(GCPtrEffSrc, bRm, 0);
+                IEMOP_HLP_DONE_DECODING_NO_LOCK_PREFIX();
                 IEM_MC_FETCH_MEM_U64(u64Target, pVCpu->iem.s.iEffSeg, GCPtrEffSrc);
                 IEM_MC_SET_RIP_U64(u64Target);
@@ -17832 +17929 @@
 {
     IEMOP_MNEMONIC("push Ev");
-    IEMOP_HLP_NO_LOCK_PREFIX(); /** @todo Too early? */
 
     /* Registers are handled by a common worker. */
@@ -17847 +17943 @@
             IEM_MC_LOCAL(RTGCPTR,   GCPtrEffSrc);
             IEM_MC_CALC_RM_EFF_ADDR(GCPtrEffSrc, bRm, 0);
+            IEMOP_HLP_DONE_DECODING_NO_LOCK_PREFIX();
             IEM_MC_FETCH_MEM_U16(u16Src, pVCpu->iem.s.iEffSeg, GCPtrEffSrc);
             IEM_MC_PUSH_U16(u16Src);
@@ -17858 +17955 @@
             IEM_MC_LOCAL(RTGCPTR,   GCPtrEffSrc);
             IEM_MC_CALC_RM_EFF_ADDR(GCPtrEffSrc, bRm, 0);
+            IEMOP_HLP_DONE_DECODING_NO_LOCK_PREFIX();
             IEM_MC_FETCH_MEM_U32(u32Src, pVCpu->iem.s.iEffSeg, GCPtrEffSrc);
             IEM_MC_PUSH_U32(u32Src);
@@ -17869 +17967 @@
             IEM_MC_LOCAL(RTGCPTR,   GCPtrEffSrc);
             IEM_MC_CALC_RM_EFF_ADDR(GCPtrEffSrc, bRm, 0);
+            IEMOP_HLP_DONE_DECODING_NO_LOCK_PREFIX();
             IEM_MC_FETCH_MEM_U64(u64Src, pVCpu->iem.s.iEffSeg, GCPtrEffSrc);
             IEM_MC_PUSH_U64(u64Src);