From 2bce7bc460b83d1da749976bb04b37055a03f602 Mon Sep 17 00:00:00 2001 From: Henrik Rydgard Date: Sun, 7 Dec 2014 23:09:38 +0100 Subject: [PATCH] X64Emitter: Merge some AVX stuff from Dolphin --- Common/x64Emitter.cpp | 79 ++++++++++++++++++++++++++++++++++++++++--- Common/x64Emitter.h | 78 ++++++++++++++++++++++++++++++++++++++++-- 2 files changed, 150 insertions(+), 7 deletions(-) diff --git a/Common/x64Emitter.cpp b/Common/x64Emitter.cpp index 3e8f3e958b..d0b376d460 100644 --- a/Common/x64Emitter.cpp +++ b/Common/x64Emitter.cpp @@ -1838,14 +1838,85 @@ void XEmitter::VSUBPD(X64Reg regOp1, X64Reg regOp2, OpArg arg) {WriteAVXOp(0x6 void XEmitter::VMULPD(X64Reg regOp1, X64Reg regOp2, OpArg arg) {WriteAVXOp(0x66, sseMUL, regOp1, regOp2, arg);} void XEmitter::VDIVPD(X64Reg regOp1, X64Reg regOp2, OpArg arg) {WriteAVXOp(0x66, sseDIV, regOp1, regOp2, arg);} void XEmitter::VSQRTSD(X64Reg regOp1, X64Reg regOp2, OpArg arg) {WriteAVXOp(0xF2, sseSQRT, regOp1, regOp2, arg);} -void XEmitter::VPAND(X64Reg regOp1, X64Reg regOp2, OpArg arg) {WriteAVXOp(0x66, sseAND, regOp1, regOp2, arg);} -void XEmitter::VPANDN(X64Reg regOp1, X64Reg regOp2, OpArg arg) {WriteAVXOp(0x66, sseANDN, regOp1, regOp2, arg);} -void XEmitter::VPOR(X64Reg regOp1, X64Reg regOp2, OpArg arg) {WriteAVXOp(0x66, sseOR, regOp1, regOp2, arg);} -void XEmitter::VPXOR(X64Reg regOp1, X64Reg regOp2, OpArg arg) {WriteAVXOp(0x66, sseXOR, regOp1, regOp2, arg);} void XEmitter::VSHUFPD(X64Reg regOp1, X64Reg regOp2, OpArg arg, u8 shuffle) {WriteAVXOp(0x66, sseSHUF, regOp1, regOp2, arg, 1); Write8(shuffle);} void XEmitter::VUNPCKLPD(X64Reg regOp1, X64Reg regOp2, OpArg arg){WriteAVXOp(0x66, 0x14, regOp1, regOp2, arg);} void XEmitter::VUNPCKHPD(X64Reg regOp1, X64Reg regOp2, OpArg arg){WriteAVXOp(0x66, 0x15, regOp1, regOp2, arg);} +void XEmitter::VANDPS(X64Reg regOp1, X64Reg regOp2, OpArg arg) { WriteAVXOp(0x00, sseAND, regOp1, regOp2, arg); } +void XEmitter::VANDPD(X64Reg regOp1, X64Reg regOp2, OpArg arg) { WriteAVXOp(0x66, sseAND, regOp1, regOp2, arg); } +void XEmitter::VANDNPS(X64Reg regOp1, X64Reg regOp2, OpArg arg) { WriteAVXOp(0x00, sseANDN, regOp1, regOp2, arg); } +void XEmitter::VANDNPD(X64Reg regOp1, X64Reg regOp2, OpArg arg) { WriteAVXOp(0x66, sseANDN, regOp1, regOp2, arg); } +void XEmitter::VORPS(X64Reg regOp1, X64Reg regOp2, OpArg arg) { WriteAVXOp(0x00, sseOR, regOp1, regOp2, arg); } +void XEmitter::VORPD(X64Reg regOp1, X64Reg regOp2, OpArg arg) { WriteAVXOp(0x66, sseOR, regOp1, regOp2, arg); } +void XEmitter::VXORPS(X64Reg regOp1, X64Reg regOp2, OpArg arg) { WriteAVXOp(0x00, sseXOR, regOp1, regOp2, arg); } +void XEmitter::VXORPD(X64Reg regOp1, X64Reg regOp2, OpArg arg) { WriteAVXOp(0x66, sseXOR, regOp1, regOp2, arg); } + +void XEmitter::VPAND(X64Reg regOp1, X64Reg regOp2, OpArg arg) { WriteAVXOp(0x66, 0xDB, regOp1, regOp2, arg); } +void XEmitter::VPANDN(X64Reg regOp1, X64Reg regOp2, OpArg arg) { WriteAVXOp(0x66, 0xDF, regOp1, regOp2, arg); } +void XEmitter::VPOR(X64Reg regOp1, X64Reg regOp2, OpArg arg) { WriteAVXOp(0x66, 0xEB, regOp1, regOp2, arg); } +void XEmitter::VPXOR(X64Reg regOp1, X64Reg regOp2, OpArg arg) { WriteAVXOp(0x66, 0xEF, regOp1, regOp2, arg); } + +void XEmitter::VFMADD132PS(X64Reg regOp1, X64Reg regOp2, OpArg arg) { WriteAVXOp(0x66, 0x3898, regOp1, regOp2, arg); } +void XEmitter::VFMADD213PS(X64Reg regOp1, X64Reg regOp2, OpArg arg) { WriteAVXOp(0x66, 0x38A8, regOp1, regOp2, arg); } +void XEmitter::VFMADD231PS(X64Reg regOp1, X64Reg regOp2, OpArg arg) { WriteAVXOp(0x66, 0x38B8, regOp1, regOp2, arg); } +void XEmitter::VFMADD132PD(X64Reg regOp1, X64Reg regOp2, OpArg arg) { WriteAVXOp(0x66, 0x3898, regOp1, regOp2, arg, 1); } +void XEmitter::VFMADD213PD(X64Reg regOp1, X64Reg regOp2, OpArg arg) { WriteAVXOp(0x66, 0x38A8, regOp1, regOp2, arg, 1); } +void XEmitter::VFMADD231PD(X64Reg regOp1, X64Reg regOp2, OpArg arg) { WriteAVXOp(0x66, 0x38B8, regOp1, regOp2, arg, 1); } +void XEmitter::VFMADD132SS(X64Reg regOp1, X64Reg regOp2, OpArg arg) { WriteAVXOp(0x66, 0x3899, regOp1, regOp2, arg); } +void XEmitter::VFMADD213SS(X64Reg regOp1, X64Reg regOp2, OpArg arg) { WriteAVXOp(0x66, 0x38A9, regOp1, regOp2, arg); } +void XEmitter::VFMADD231SS(X64Reg regOp1, X64Reg regOp2, OpArg arg) { WriteAVXOp(0x66, 0x38B9, regOp1, regOp2, arg); } +void XEmitter::VFMADD132SD(X64Reg regOp1, X64Reg regOp2, OpArg arg) { WriteAVXOp(0x66, 0x3899, regOp1, regOp2, arg, 1); } +void XEmitter::VFMADD213SD(X64Reg regOp1, X64Reg regOp2, OpArg arg) { WriteAVXOp(0x66, 0x38A9, regOp1, regOp2, arg, 1); } +void XEmitter::VFMADD231SD(X64Reg regOp1, X64Reg regOp2, OpArg arg) { WriteAVXOp(0x66, 0x38B9, regOp1, regOp2, arg, 1); } +void XEmitter::VFMSUB132PS(X64Reg regOp1, X64Reg regOp2, OpArg arg) { WriteAVXOp(0x66, 0x389A, regOp1, regOp2, arg); } +void XEmitter::VFMSUB213PS(X64Reg regOp1, X64Reg regOp2, OpArg arg) { WriteAVXOp(0x66, 0x38AA, regOp1, regOp2, arg); } +void XEmitter::VFMSUB231PS(X64Reg regOp1, X64Reg regOp2, OpArg arg) { WriteAVXOp(0x66, 0x38BA, regOp1, regOp2, arg); } +void XEmitter::VFMSUB132PD(X64Reg regOp1, X64Reg regOp2, OpArg arg) { WriteAVXOp(0x66, 0x389A, regOp1, regOp2, arg, 1); } +void XEmitter::VFMSUB213PD(X64Reg regOp1, X64Reg regOp2, OpArg arg) { WriteAVXOp(0x66, 0x38AA, regOp1, regOp2, arg, 1); } +void XEmitter::VFMSUB231PD(X64Reg regOp1, X64Reg regOp2, OpArg arg) { WriteAVXOp(0x66, 0x38BA, regOp1, regOp2, arg, 1); } +void XEmitter::VFMSUB132SS(X64Reg regOp1, X64Reg regOp2, OpArg arg) { WriteAVXOp(0x66, 0x389B, regOp1, regOp2, arg); } +void XEmitter::VFMSUB213SS(X64Reg regOp1, X64Reg regOp2, OpArg arg) { WriteAVXOp(0x66, 0x38AB, regOp1, regOp2, arg); } +void XEmitter::VFMSUB231SS(X64Reg regOp1, X64Reg regOp2, OpArg arg) { WriteAVXOp(0x66, 0x38BB, regOp1, regOp2, arg); } +void XEmitter::VFMSUB132SD(X64Reg regOp1, X64Reg regOp2, OpArg arg) { WriteAVXOp(0x66, 0x389B, regOp1, regOp2, arg, 1); } +void XEmitter::VFMSUB213SD(X64Reg regOp1, X64Reg regOp2, OpArg arg) { WriteAVXOp(0x66, 0x38AB, regOp1, regOp2, arg, 1); } +void XEmitter::VFMSUB231SD(X64Reg regOp1, X64Reg regOp2, OpArg arg) { WriteAVXOp(0x66, 0x38BB, regOp1, regOp2, arg, 1); } +void XEmitter::VFNMADD132PS(X64Reg regOp1, X64Reg regOp2, OpArg arg) { WriteAVXOp(0x66, 0x389C, regOp1, regOp2, arg); } +void XEmitter::VFNMADD213PS(X64Reg regOp1, X64Reg regOp2, OpArg arg) { WriteAVXOp(0x66, 0x38AC, regOp1, regOp2, arg); } +void XEmitter::VFNMADD231PS(X64Reg regOp1, X64Reg regOp2, OpArg arg) { WriteAVXOp(0x66, 0x38BC, regOp1, regOp2, arg); } +void XEmitter::VFNMADD132PD(X64Reg regOp1, X64Reg regOp2, OpArg arg) { WriteAVXOp(0x66, 0x389C, regOp1, regOp2, arg, 1); } +void XEmitter::VFNMADD213PD(X64Reg regOp1, X64Reg regOp2, OpArg arg) { WriteAVXOp(0x66, 0x38AC, regOp1, regOp2, arg, 1); } +void XEmitter::VFNMADD231PD(X64Reg regOp1, X64Reg regOp2, OpArg arg) { WriteAVXOp(0x66, 0x38BC, regOp1, regOp2, arg, 1); } +void XEmitter::VFNMADD132SS(X64Reg regOp1, X64Reg regOp2, OpArg arg) { WriteAVXOp(0x66, 0x389D, regOp1, regOp2, arg); } +void XEmitter::VFNMADD213SS(X64Reg regOp1, X64Reg regOp2, OpArg arg) { WriteAVXOp(0x66, 0x38AD, regOp1, regOp2, arg); } +void XEmitter::VFNMADD231SS(X64Reg regOp1, X64Reg regOp2, OpArg arg) { WriteAVXOp(0x66, 0x38BD, regOp1, regOp2, arg); } +void XEmitter::VFNMADD132SD(X64Reg regOp1, X64Reg regOp2, OpArg arg) { WriteAVXOp(0x66, 0x389D, regOp1, regOp2, arg, 1); } +void XEmitter::VFNMADD213SD(X64Reg regOp1, X64Reg regOp2, OpArg arg) { WriteAVXOp(0x66, 0x38AD, regOp1, regOp2, arg, 1); } +void XEmitter::VFNMADD231SD(X64Reg regOp1, X64Reg regOp2, OpArg arg) { WriteAVXOp(0x66, 0x38BD, regOp1, regOp2, arg, 1); } +void XEmitter::VFNMSUB132PS(X64Reg regOp1, X64Reg regOp2, OpArg arg) { WriteAVXOp(0x66, 0x389E, regOp1, regOp2, arg); } +void XEmitter::VFNMSUB213PS(X64Reg regOp1, X64Reg regOp2, OpArg arg) { WriteAVXOp(0x66, 0x38AE, regOp1, regOp2, arg); } +void XEmitter::VFNMSUB231PS(X64Reg regOp1, X64Reg regOp2, OpArg arg) { WriteAVXOp(0x66, 0x38BE, regOp1, regOp2, arg); } +void XEmitter::VFNMSUB132PD(X64Reg regOp1, X64Reg regOp2, OpArg arg) { WriteAVXOp(0x66, 0x389E, regOp1, regOp2, arg, 1); } +void XEmitter::VFNMSUB213PD(X64Reg regOp1, X64Reg regOp2, OpArg arg) { WriteAVXOp(0x66, 0x38AE, regOp1, regOp2, arg, 1); } +void XEmitter::VFNMSUB231PD(X64Reg regOp1, X64Reg regOp2, OpArg arg) { WriteAVXOp(0x66, 0x38BE, regOp1, regOp2, arg, 1); } +void XEmitter::VFNMSUB132SS(X64Reg regOp1, X64Reg regOp2, OpArg arg) { WriteAVXOp(0x66, 0x389F, regOp1, regOp2, arg); } +void XEmitter::VFNMSUB213SS(X64Reg regOp1, X64Reg regOp2, OpArg arg) { WriteAVXOp(0x66, 0x38AF, regOp1, regOp2, arg); } +void XEmitter::VFNMSUB231SS(X64Reg regOp1, X64Reg regOp2, OpArg arg) { WriteAVXOp(0x66, 0x38BF, regOp1, regOp2, arg); } +void XEmitter::VFNMSUB132SD(X64Reg regOp1, X64Reg regOp2, OpArg arg) { WriteAVXOp(0x66, 0x389F, regOp1, regOp2, arg, 1); } +void XEmitter::VFNMSUB213SD(X64Reg regOp1, X64Reg regOp2, OpArg arg) { WriteAVXOp(0x66, 0x38AF, regOp1, regOp2, arg, 1); } +void XEmitter::VFNMSUB231SD(X64Reg regOp1, X64Reg regOp2, OpArg arg) { WriteAVXOp(0x66, 0x38BF, regOp1, regOp2, arg, 1); } +void XEmitter::VFMADDSUB132PS(X64Reg regOp1, X64Reg regOp2, OpArg arg) { WriteAVXOp(0x66, 0x3896, regOp1, regOp2, arg); } +void XEmitter::VFMADDSUB213PS(X64Reg regOp1, X64Reg regOp2, OpArg arg) { WriteAVXOp(0x66, 0x38A6, regOp1, regOp2, arg); } +void XEmitter::VFMADDSUB231PS(X64Reg regOp1, X64Reg regOp2, OpArg arg) { WriteAVXOp(0x66, 0x38B6, regOp1, regOp2, arg); } +void XEmitter::VFMADDSUB132PD(X64Reg regOp1, X64Reg regOp2, OpArg arg) { WriteAVXOp(0x66, 0x3896, regOp1, regOp2, arg, 1); } +void XEmitter::VFMADDSUB213PD(X64Reg regOp1, X64Reg regOp2, OpArg arg) { WriteAVXOp(0x66, 0x38A6, regOp1, regOp2, arg, 1); } +void XEmitter::VFMADDSUB231PD(X64Reg regOp1, X64Reg regOp2, OpArg arg) { WriteAVXOp(0x66, 0x38B6, regOp1, regOp2, arg, 1); } +void XEmitter::VFMSUBADD132PS(X64Reg regOp1, X64Reg regOp2, OpArg arg) { WriteAVXOp(0x66, 0x3897, regOp1, regOp2, arg); } +void XEmitter::VFMSUBADD213PS(X64Reg regOp1, X64Reg regOp2, OpArg arg) { WriteAVXOp(0x66, 0x38A7, regOp1, regOp2, arg); } +void XEmitter::VFMSUBADD231PS(X64Reg regOp1, X64Reg regOp2, OpArg arg) { WriteAVXOp(0x66, 0x38B7, regOp1, regOp2, arg); } +void XEmitter::VFMSUBADD132PD(X64Reg regOp1, X64Reg regOp2, OpArg arg) { WriteAVXOp(0x66, 0x3897, regOp1, regOp2, arg, 1); } +void XEmitter::VFMSUBADD213PD(X64Reg regOp1, X64Reg regOp2, OpArg arg) { WriteAVXOp(0x66, 0x38A7, regOp1, regOp2, arg, 1); } +void XEmitter::VFMSUBADD231PD(X64Reg regOp1, X64Reg regOp2, OpArg arg) { WriteAVXOp(0x66, 0x38B7, regOp1, regOp2, arg, 1); } + void XEmitter::SARX(int bits, X64Reg regOp1, OpArg arg, X64Reg regOp2) {WriteBMI2Op(bits, 0xF3, 0x38F7, regOp1, regOp2, arg);} void XEmitter::SHLX(int bits, X64Reg regOp1, OpArg arg, X64Reg regOp2) {WriteBMI2Op(bits, 0x66, 0x38F7, regOp1, regOp2, arg);} void XEmitter::SHRX(int bits, X64Reg regOp1, OpArg arg, X64Reg regOp2) {WriteBMI2Op(bits, 0xF2, 0x38F7, regOp1, regOp2, arg);} diff --git a/Common/x64Emitter.h b/Common/x64Emitter.h index 838cf94e9b..fb1c2c5027 100644 --- a/Common/x64Emitter.h +++ b/Common/x64Emitter.h @@ -872,13 +872,85 @@ public: void VMULPD(X64Reg regOp1, X64Reg regOp2, OpArg arg); void VDIVPD(X64Reg regOp1, X64Reg regOp2, OpArg arg); void VSQRTSD(X64Reg regOp1, X64Reg regOp2, OpArg arg); + void VSHUFPD(X64Reg regOp1, X64Reg regOp2, OpArg arg, u8 shuffle); + void VUNPCKLPD(X64Reg regOp1, X64Reg regOp2, OpArg arg); + void VUNPCKHPD(X64Reg regOp1, X64Reg regOp2, OpArg arg); + + void VANDPS(X64Reg regOp1, X64Reg regOp2, OpArg arg); + void VANDPD(X64Reg regOp1, X64Reg regOp2, OpArg arg); + void VANDNPS(X64Reg regOp1, X64Reg regOp2, OpArg arg); + void VANDNPD(X64Reg regOp1, X64Reg regOp2, OpArg arg); + void VORPS(X64Reg regOp1, X64Reg regOp2, OpArg arg); + void VORPD(X64Reg regOp1, X64Reg regOp2, OpArg arg); + void VXORPS(X64Reg regOp1, X64Reg regOp2, OpArg arg); + void VXORPD(X64Reg regOp1, X64Reg regOp2, OpArg arg); + void VPAND(X64Reg regOp1, X64Reg regOp2, OpArg arg); void VPANDN(X64Reg regOp1, X64Reg regOp2, OpArg arg); void VPOR(X64Reg regOp1, X64Reg regOp2, OpArg arg); void VPXOR(X64Reg regOp1, X64Reg regOp2, OpArg arg); - void VSHUFPD(X64Reg regOp1, X64Reg regOp2, OpArg arg, u8 shuffle); - void VUNPCKLPD(X64Reg regOp1, X64Reg regOp2, OpArg arg); - void VUNPCKHPD(X64Reg regOp1, X64Reg regOp2, OpArg arg); + + // FMA3 + void VFMADD132PS(X64Reg regOp1, X64Reg regOp2, OpArg arg); + void VFMADD213PS(X64Reg regOp1, X64Reg regOp2, OpArg arg); + void VFMADD231PS(X64Reg regOp1, X64Reg regOp2, OpArg arg); + void VFMADD132PD(X64Reg regOp1, X64Reg regOp2, OpArg arg); + void VFMADD213PD(X64Reg regOp1, X64Reg regOp2, OpArg arg); + void VFMADD231PD(X64Reg regOp1, X64Reg regOp2, OpArg arg); + void VFMADD132SS(X64Reg regOp1, X64Reg regOp2, OpArg arg); + void VFMADD213SS(X64Reg regOp1, X64Reg regOp2, OpArg arg); + void VFMADD231SS(X64Reg regOp1, X64Reg regOp2, OpArg arg); + void VFMADD132SD(X64Reg regOp1, X64Reg regOp2, OpArg arg); + void VFMADD213SD(X64Reg regOp1, X64Reg regOp2, OpArg arg); + void VFMADD231SD(X64Reg regOp1, X64Reg regOp2, OpArg arg); + void VFMSUB132PS(X64Reg regOp1, X64Reg regOp2, OpArg arg); + void VFMSUB213PS(X64Reg regOp1, X64Reg regOp2, OpArg arg); + void VFMSUB231PS(X64Reg regOp1, X64Reg regOp2, OpArg arg); + void VFMSUB132PD(X64Reg regOp1, X64Reg regOp2, OpArg arg); + void VFMSUB213PD(X64Reg regOp1, X64Reg regOp2, OpArg arg); + void VFMSUB231PD(X64Reg regOp1, X64Reg regOp2, OpArg arg); + void VFMSUB132SS(X64Reg regOp1, X64Reg regOp2, OpArg arg); + void VFMSUB213SS(X64Reg regOp1, X64Reg regOp2, OpArg arg); + void VFMSUB231SS(X64Reg regOp1, X64Reg regOp2, OpArg arg); + void VFMSUB132SD(X64Reg regOp1, X64Reg regOp2, OpArg arg); + void VFMSUB213SD(X64Reg regOp1, X64Reg regOp2, OpArg arg); + void VFMSUB231SD(X64Reg regOp1, X64Reg regOp2, OpArg arg); + void VFNMADD132PS(X64Reg regOp1, X64Reg regOp2, OpArg arg); + void VFNMADD213PS(X64Reg regOp1, X64Reg regOp2, OpArg arg); + void VFNMADD231PS(X64Reg regOp1, X64Reg regOp2, OpArg arg); + void VFNMADD132PD(X64Reg regOp1, X64Reg regOp2, OpArg arg); + void VFNMADD213PD(X64Reg regOp1, X64Reg regOp2, OpArg arg); + void VFNMADD231PD(X64Reg regOp1, X64Reg regOp2, OpArg arg); + void VFNMADD132SS(X64Reg regOp1, X64Reg regOp2, OpArg arg); + void VFNMADD213SS(X64Reg regOp1, X64Reg regOp2, OpArg arg); + void VFNMADD231SS(X64Reg regOp1, X64Reg regOp2, OpArg arg); + void VFNMADD132SD(X64Reg regOp1, X64Reg regOp2, OpArg arg); + void VFNMADD213SD(X64Reg regOp1, X64Reg regOp2, OpArg arg); + void VFNMADD231SD(X64Reg regOp1, X64Reg regOp2, OpArg arg); + void VFNMSUB132PS(X64Reg regOp1, X64Reg regOp2, OpArg arg); + void VFNMSUB213PS(X64Reg regOp1, X64Reg regOp2, OpArg arg); + void VFNMSUB231PS(X64Reg regOp1, X64Reg regOp2, OpArg arg); + void VFNMSUB132PD(X64Reg regOp1, X64Reg regOp2, OpArg arg); + void VFNMSUB213PD(X64Reg regOp1, X64Reg regOp2, OpArg arg); + void VFNMSUB231PD(X64Reg regOp1, X64Reg regOp2, OpArg arg); + void VFNMSUB132SS(X64Reg regOp1, X64Reg regOp2, OpArg arg); + void VFNMSUB213SS(X64Reg regOp1, X64Reg regOp2, OpArg arg); + void VFNMSUB231SS(X64Reg regOp1, X64Reg regOp2, OpArg arg); + void VFNMSUB132SD(X64Reg regOp1, X64Reg regOp2, OpArg arg); + void VFNMSUB213SD(X64Reg regOp1, X64Reg regOp2, OpArg arg); + void VFNMSUB231SD(X64Reg regOp1, X64Reg regOp2, OpArg arg); + void VFMADDSUB132PS(X64Reg regOp1, X64Reg regOp2, OpArg arg); + void VFMADDSUB213PS(X64Reg regOp1, X64Reg regOp2, OpArg arg); + void VFMADDSUB231PS(X64Reg regOp1, X64Reg regOp2, OpArg arg); + void VFMADDSUB132PD(X64Reg regOp1, X64Reg regOp2, OpArg arg); + void VFMADDSUB213PD(X64Reg regOp1, X64Reg regOp2, OpArg arg); + void VFMADDSUB231PD(X64Reg regOp1, X64Reg regOp2, OpArg arg); + void VFMSUBADD132PS(X64Reg regOp1, X64Reg regOp2, OpArg arg); + void VFMSUBADD213PS(X64Reg regOp1, X64Reg regOp2, OpArg arg); + void VFMSUBADD231PS(X64Reg regOp1, X64Reg regOp2, OpArg arg); + void VFMSUBADD132PD(X64Reg regOp1, X64Reg regOp2, OpArg arg); + void VFMSUBADD213PD(X64Reg regOp1, X64Reg regOp2, OpArg arg); + void VFMSUBADD231PD(X64Reg regOp1, X64Reg regOp2, OpArg arg); // VEX GPR instructions void SARX(int bits, X64Reg regOp1, OpArg arg, X64Reg regOp2);