// FMA3 - Intel 3 operand Fused Multiply-Add instructions
//===----------------------------------------------------------------------===//
+let Constraints = "$src1 = $dst" in {
multiclass fma3p_rm<bits<8> opc, string OpcodeStr> {
+let neverHasSideEffects = 1 in {
def r : FMA3<opc, MRMSrcReg, (outs VR128:$dst),
- (ins VR128:$src1, VR128:$src2),
- !strconcat(OpcodeStr, "\t{$src2, $src1, $dst|$dst, $src1, $src2}"),
+ (ins VR128:$src1, VR128:$src2, VR128:$src3),
+ !strconcat(OpcodeStr, "\t{$src3, $src2, $dst|$dst, $src2, $src3}"),
[]>;
+ let mayLoad = 1 in
def m : FMA3<opc, MRMSrcMem, (outs VR128:$dst),
- (ins VR128:$src1, f128mem:$src2),
- !strconcat(OpcodeStr, "\t{$src2, $src1, $dst|$dst, $src1, $src2}"),
+ (ins VR128:$src1, VR128:$src2, f128mem:$src3),
+ !strconcat(OpcodeStr, "\t{$src3, $src2, $dst|$dst, $src2, $src3}"),
[]>;
def rY : FMA3<opc, MRMSrcReg, (outs VR256:$dst),
- (ins VR256:$src1, VR256:$src2),
- !strconcat(OpcodeStr, "\t{$src2, $src1, $dst|$dst, $src1, $src2}"),
+ (ins VR256:$src1, VR256:$src2, VR256:$src3),
+ !strconcat(OpcodeStr, "\t{$src3, $src2, $dst|$dst, $src2, $src3}"),
[]>;
+ let mayLoad = 1 in
def mY : FMA3<opc, MRMSrcMem, (outs VR256:$dst),
- (ins VR256:$src1, f256mem:$src2),
- !strconcat(OpcodeStr, "\t{$src2, $src1, $dst|$dst, $src1, $src2}"),
+ (ins VR256:$src1, VR256:$src2, f256mem:$src3),
+ !strconcat(OpcodeStr, "\t{$src3, $src2, $dst|$dst, $src2, $src3}"),
[]>;
+} // neverHasSideEffects = 1
+}
+
+// Intrinsic for 132 pattern
+multiclass fma3p_rm_int<bits<8> opc, string OpcodeStr,
+ PatFrag MemFrag128, PatFrag MemFrag256,
+ Intrinsic Int128, Intrinsic Int256> {
+ def r_Int : FMA3<opc, MRMSrcReg, (outs VR128:$dst),
+ (ins VR128:$src1, VR128:$src2, VR128:$src3),
+ !strconcat(OpcodeStr, "\t{$src3, $src2, $dst|$dst, $src2, $src3}"),
+ [(set VR128:$dst, (Int128 VR128:$src1, VR128:$src3, VR128:$src2))]>;
+ def m_Int : FMA3<opc, MRMSrcMem, (outs VR128:$dst),
+ (ins VR128:$src1, VR128:$src2, f128mem:$src3),
+ !strconcat(OpcodeStr, "\t{$src3, $src2, $dst|$dst, $src2, $src3}"),
+ [(set VR128:$dst,
+ (Int128 VR128:$src1, (MemFrag128 addr:$src3), VR128:$src2))]>;
+ def rY_Int : FMA3<opc, MRMSrcReg, (outs VR256:$dst),
+ (ins VR256:$src1, VR256:$src2, VR256:$src3),
+ !strconcat(OpcodeStr, "\t{$src3, $src2, $dst|$dst, $src2, $src3}"),
+ [(set VR256:$dst, (Int256 VR256:$src1, VR256:$src3, VR256:$src2))]>;
+ def mY_Int : FMA3<opc, MRMSrcMem, (outs VR256:$dst),
+ (ins VR256:$src1, VR256:$src2, f256mem:$src3),
+ !strconcat(OpcodeStr, "\t{$src3, $src2, $dst|$dst, $src2, $src3}"),
+ [(set VR256:$dst,
+ (Int256 VR256:$src1, (MemFrag256 addr:$src3), VR256:$src2))]>;
}
+} // Constraints = "$src1 = $dst"
multiclass fma3p_forms<bits<8> opc132, bits<8> opc213, bits<8> opc231,
- string OpcodeStr, string PackTy> {
- defm r132 : fma3p_rm<opc132, !strconcat(OpcodeStr, !strconcat("132", PackTy))>;
- defm r213 : fma3p_rm<opc213, !strconcat(OpcodeStr, !strconcat("213", PackTy))>;
- defm r231 : fma3p_rm<opc231, !strconcat(OpcodeStr, !strconcat("231", PackTy))>;
+ string OpcodeStr, string PackTy,
+ PatFrag MemFrag128, PatFrag MemFrag256,
+ Intrinsic Int128, Intrinsic Int256> {
+ defm r132 : fma3p_rm_int <opc132, !strconcat(OpcodeStr,
+ !strconcat("132", PackTy)), MemFrag128, MemFrag256,
+ Int128, Int256>;
+ defm r132 : fma3p_rm <opc132, !strconcat(OpcodeStr, !strconcat("132", PackTy))>;
+ defm r213 : fma3p_rm <opc213, !strconcat(OpcodeStr, !strconcat("213", PackTy))>;
+ defm r231 : fma3p_rm <opc231, !strconcat(OpcodeStr, !strconcat("231", PackTy))>;
}
// Fused Multiply-Add
let ExeDomain = SSEPackedSingle in {
- defm VFMADDPS : fma3p_forms<0x98, 0xA8, 0xB8, "vfmadd", "ps">;
- defm VFMSUBPS : fma3p_forms<0x9A, 0xAA, 0xBA, "vfmsub", "ps">;
- defm VFMADDSUBPS : fma3p_forms<0x96, 0xA6, 0xB6, "vfmaddsub", "ps">;
- defm VFMSUBADDPS : fma3p_forms<0x97, 0xA7, 0xB7, "vfmsubadd", "ps">;
+ defm VFMADDPS : fma3p_forms<0x98, 0xA8, 0xB8, "vfmadd", "ps", memopv4f32,
+ memopv8f32, int_x86_fma_vfmadd_ps, int_x86_fma_vfmadd_ps_256>;
+ defm VFMSUBPS : fma3p_forms<0x9A, 0xAA, 0xBA, "vfmsub", "ps", memopv4f32,
+ memopv8f32, int_x86_fma_vfmsub_ps, int_x86_fma_vfmsub_ps_256>;
+ defm VFMADDSUBPS : fma3p_forms<0x96, 0xA6, 0xB6, "vfmaddsub", "ps",
+ memopv4f32, memopv8f32, int_x86_fma_vfmaddsub_ps,
+ int_x86_fma_vfmaddsub_ps_256>;
+ defm VFMSUBADDPS : fma3p_forms<0x97, 0xA7, 0xB7, "vfmsubadd", "ps",
+ memopv4f32, memopv8f32, int_x86_fma_vfmsubadd_ps,
+ int_x86_fma_vfmaddsub_ps_256>;
}
let ExeDomain = SSEPackedDouble in {
- defm VFMADDPD : fma3p_forms<0x98, 0xA8, 0xB8, "vfmadd", "pd">, VEX_W;
- defm VFMSUBPD : fma3p_forms<0x9A, 0xAA, 0xBA, "vfmsub", "pd">, VEX_W;
- defm VFMADDSUBPD : fma3p_forms<0x96, 0xA6, 0xB6, "vfmaddsub", "pd">, VEX_W;
- defm VFMSUBADDPD : fma3p_forms<0x97, 0xA7, 0xB7, "vfmsubadd", "pd">, VEX_W;
+ defm VFMADDPD : fma3p_forms<0x98, 0xA8, 0xB8, "vfmadd", "pd", memopv2f64,
+ memopv4f64, int_x86_fma_vfmadd_pd, int_x86_fma_vfmadd_pd_256>, VEX_W;
+ defm VFMSUBPD : fma3p_forms<0x9A, 0xAA, 0xBA, "vfmsub", "pd", memopv2f64,
+ memopv4f64, int_x86_fma_vfmsub_pd, int_x86_fma_vfmsub_pd_256>, VEX_W;
+ defm VFMADDSUBPD : fma3p_forms<0x96, 0xA6, 0xB6, "vfmaddsub", "pd", memopv2f64,
+ memopv4f64, int_x86_fma_vfmaddsub_pd, int_x86_fma_vfmaddsub_pd_256>, VEX_W;
+ defm VFMSUBADDPD : fma3p_forms<0x97, 0xA7, 0xB7, "vfmsubadd", "pd", memopv2f64,
+ memopv4f64, int_x86_fma_vfmsubadd_pd, int_x86_fma_vfmsubadd_pd_256>, VEX_W;
}
// Fused Negative Multiply-Add
let ExeDomain = SSEPackedSingle in {
- defm VFNMADDPS : fma3p_forms<0x9C, 0xAC, 0xBC, "vfnmadd", "ps">;
- defm VFNMSUBPS : fma3p_forms<0x9E, 0xAE, 0xBE, "vfnmsub", "ps">;
+ defm VFNMADDPS : fma3p_forms<0x9C, 0xAC, 0xBC, "vfnmadd", "ps", memopv4f32,
+ memopv8f32, int_x86_fma_vfnmadd_ps, int_x86_fma_vfnmadd_ps_256>;
+ defm VFNMSUBPS : fma3p_forms<0x9E, 0xAE, 0xBE, "vfnmsub", "ps", memopv4f32,
+ memopv8f32, int_x86_fma_vfnmsub_ps, int_x86_fma_vfnmsub_ps_256>;
}
let ExeDomain = SSEPackedDouble in {
- defm VFNMADDPD : fma3p_forms<0x9C, 0xAC, 0xBC, "vfnmadd", "pd">, VEX_W;
- defm VFNMSUBPD : fma3p_forms<0x9E, 0xAE, 0xBE, "vfnmsub", "pd">, VEX_W;
+ defm VFNMADDPD : fma3p_forms<0x9C, 0xAC, 0xBC, "vfnmadd", "pd", memopv2f64,
+ memopv4f64, int_x86_fma_vfnmadd_pd, int_x86_fma_vfnmadd_pd_256>, VEX_W;
+ defm VFNMSUBPD : fma3p_forms<0x9E, 0xAE, 0xBE, "vfnmsub", "pd", memopv2f64,
+ memopv4f64, int_x86_fma_vfnmsub_pd, int_x86_fma_vfnmsub_pd_256>, VEX_W;
}
-multiclass fma3s_rm<bits<8> opc, string OpcodeStr, X86MemOperand x86memop> {
- def r : FMA3<opc, MRMSrcReg, (outs VR128:$dst),
- (ins VR128:$src1, VR128:$src2),
- !strconcat(OpcodeStr, "\t{$src2, $src1, $dst|$dst, $src1, $src2}"),
+
+let Constraints = "$src1 = $dst" in {
+multiclass fma3s_rm<bits<8> opc, string OpcodeStr, X86MemOperand x86memop,
+ RegisterClass RC> {
+let neverHasSideEffects = 1 in {
+ def r : FMA3<opc, MRMSrcReg, (outs RC:$dst),
+ (ins RC:$src1, RC:$src2, RC:$src3),
+ !strconcat(OpcodeStr, "\t{$src3, $src2, $dst|$dst, $src2, $src3}"),
[]>;
- def m : FMA3<opc, MRMSrcMem, (outs VR128:$dst),
- (ins VR128:$src1, x86memop:$src2),
- !strconcat(OpcodeStr, "\t{$src2, $src1, $dst|$dst, $src1, $src2}"),
+ let mayLoad = 1 in
+ def m : FMA3<opc, MRMSrcMem, (outs RC:$dst),
+ (ins RC:$src1, RC:$src2, x86memop:$src3),
+ !strconcat(OpcodeStr, "\t{$src3, $src2, $dst|$dst, $src2, $src3}"),
[]>;
+} // neverHasSideEffects = 1
}
+multiclass fma3s_rm_int<bits<8> opc, string OpcodeStr, Operand memop,
+ ComplexPattern mem_cpat, Intrinsic IntId> {
+ def r_Int : FMA3<opc, MRMSrcReg, (outs VR128:$dst),
+ (ins VR128:$src1, VR128:$src2, VR128:$src3),
+ !strconcat(OpcodeStr, "\t{$src3, $src2, $dst|$dst, $src2, $src3}"),
+ [(set VR128:$dst, (IntId VR128:$src1, VR128:$src3, VR128:$src2))]>;
+ def m_Int : FMA3<opc, MRMSrcMem, (outs VR128:$dst),
+ (ins VR128:$src1, VR128:$src2, memop:$src3),
+ !strconcat(OpcodeStr, "\t{$src3, $src2, $dst|$dst, $src2, $src3}"),
+ [(set VR128:$dst,
+ (IntId VR128:$src1, mem_cpat:$src3, VR128:$src2))]>;
+}
+} // Constraints = "$src1 = $dst"
+
multiclass fma3s_forms<bits<8> opc132, bits<8> opc213, bits<8> opc231,
- string OpcodeStr> {
- defm SSr132 : fma3s_rm<opc132, !strconcat(OpcodeStr, "132ss"), f32mem>;
- defm SSr213 : fma3s_rm<opc213, !strconcat(OpcodeStr, "213ss"), f32mem>;
- defm SSr231 : fma3s_rm<opc231, !strconcat(OpcodeStr, "231ss"), f32mem>;
- defm SDr132 : fma3s_rm<opc132, !strconcat(OpcodeStr, "132sd"), f64mem>, VEX_W;
- defm SDr213 : fma3s_rm<opc213, !strconcat(OpcodeStr, "213sd"), f64mem>, VEX_W;
- defm SDr231 : fma3s_rm<opc231, !strconcat(OpcodeStr, "231sd"), f64mem>, VEX_W;
+ string OpStr, Intrinsic IntF32, Intrinsic IntF64> {
+ defm SSr132 : fma3s_rm<opc132, !strconcat(OpStr, "132ss"), f32mem, FR32>;
+ defm SSr213 : fma3s_rm<opc213, !strconcat(OpStr, "213ss"), f32mem, FR32>;
+ defm SSr231 : fma3s_rm<opc231, !strconcat(OpStr, "231ss"), f32mem, FR32>;
+ defm SDr132 : fma3s_rm<opc132, !strconcat(OpStr, "132sd"), f64mem, FR64>, VEX_W;
+ defm SDr213 : fma3s_rm<opc213, !strconcat(OpStr, "213sd"), f64mem, FR64>, VEX_W;
+ defm SDr231 : fma3s_rm<opc231, !strconcat(OpStr, "231sd"), f64mem, FR64>, VEX_W;
+ defm SSr132 : fma3s_rm_int <opc132, !strconcat(OpStr, "132ss"), ssmem,
+ sse_load_f32, IntF32>;
+ defm SDr132 : fma3s_rm_int <opc132, !strconcat(OpStr, "132sd"), sdmem,
+ sse_load_f64, IntF64>;
}
-defm VFMADD : fma3s_forms<0x99, 0xA9, 0xB9, "vfmadd">, VEX_LIG;
-defm VFMSUB : fma3s_forms<0x9B, 0xAB, 0xBB, "vfmsub">, VEX_LIG;
+defm VFMADD : fma3s_forms<0x99, 0xA9, 0xB9, "vfmadd", int_x86_fma_vfmadd_ss,
+ int_x86_fma_vfmadd_sd>, VEX_LIG;
+defm VFMSUB : fma3s_forms<0x9B, 0xAB, 0xBB, "vfmsub", int_x86_fma_vfmsub_ss,
+ int_x86_fma_vfmsub_sd>, VEX_LIG;
+
+defm VFNMADD : fma3s_forms<0x9D, 0xAD, 0xBD, "vfnmadd", int_x86_fma_vfnmadd_ss,
+ int_x86_fma_vfnmadd_sd>, VEX_LIG;
+defm VFNMSUB : fma3s_forms<0x9F, 0xAF, 0xBF, "vfnmsub", int_x86_fma_vfnmsub_ss,
+ int_x86_fma_vfnmsub_sd>, VEX_LIG;
-defm VFNMADD : fma3s_forms<0x9D, 0xAD, 0xBD, "vfnmadd">, VEX_LIG;
-defm VFNMSUB : fma3s_forms<0x9F, 0xAF, 0xBF, "vfnmsub">, VEX_LIG;
//===----------------------------------------------------------------------===//
// FMA4 - AMD 4 operand Fused Multiply-Add instructions
} // isCodeGenOnly = 1
}
+let Predicates = [HasFMA4] in {
+
defm VFMADDSS4 : fma4s<0x6A, "vfmaddss", ssmem, sse_load_f32,
- int_x86_fma4_vfmadd_ss>;
+ int_x86_fma_vfmadd_ss>;
defm VFMADDSD4 : fma4s<0x6B, "vfmaddsd", sdmem, sse_load_f64,
- int_x86_fma4_vfmadd_sd>;
-defm VFMADDPS4 : fma4p<0x68, "vfmaddps", int_x86_fma4_vfmadd_ps,
- int_x86_fma4_vfmadd_ps_256, memopv4f32, memopv8f32>;
-defm VFMADDPD4 : fma4p<0x69, "vfmaddpd", int_x86_fma4_vfmadd_pd,
- int_x86_fma4_vfmadd_pd_256, memopv2f64, memopv4f64>;
+ int_x86_fma_vfmadd_sd>;
+defm VFMADDPS4 : fma4p<0x68, "vfmaddps", int_x86_fma_vfmadd_ps,
+ int_x86_fma_vfmadd_ps_256, memopv4f32, memopv8f32>;
+defm VFMADDPD4 : fma4p<0x69, "vfmaddpd", int_x86_fma_vfmadd_pd,
+ int_x86_fma_vfmadd_pd_256, memopv2f64, memopv4f64>;
defm VFMSUBSS4 : fma4s<0x6E, "vfmsubss", ssmem, sse_load_f32,
- int_x86_fma4_vfmsub_ss>;
+ int_x86_fma_vfmsub_ss>;
defm VFMSUBSD4 : fma4s<0x6F, "vfmsubsd", sdmem, sse_load_f64,
- int_x86_fma4_vfmsub_sd>;
-defm VFMSUBPS4 : fma4p<0x6C, "vfmsubps", int_x86_fma4_vfmsub_ps,
- int_x86_fma4_vfmsub_ps_256, memopv4f32, memopv8f32>;
-defm VFMSUBPD4 : fma4p<0x6D, "vfmsubpd", int_x86_fma4_vfmsub_pd,
- int_x86_fma4_vfmsub_pd_256, memopv2f64, memopv4f64>;
+ int_x86_fma_vfmsub_sd>;
+defm VFMSUBPS4 : fma4p<0x6C, "vfmsubps", int_x86_fma_vfmsub_ps,
+ int_x86_fma_vfmsub_ps_256, memopv4f32, memopv8f32>;
+defm VFMSUBPD4 : fma4p<0x6D, "vfmsubpd", int_x86_fma_vfmsub_pd,
+ int_x86_fma_vfmsub_pd_256, memopv2f64, memopv4f64>;
defm VFNMADDSS4 : fma4s<0x7A, "vfnmaddss", ssmem, sse_load_f32,
- int_x86_fma4_vfnmadd_ss>;
+ int_x86_fma_vfnmadd_ss>;
defm VFNMADDSD4 : fma4s<0x7B, "vfnmaddsd", sdmem, sse_load_f64,
- int_x86_fma4_vfnmadd_sd>;
-defm VFNMADDPS4 : fma4p<0x78, "vfnmaddps", int_x86_fma4_vfnmadd_ps,
- int_x86_fma4_vfnmadd_ps_256, memopv4f32, memopv8f32>;
-defm VFNMADDPD4 : fma4p<0x79, "vfnmaddpd", int_x86_fma4_vfnmadd_pd,
- int_x86_fma4_vfnmadd_pd_256, memopv2f64, memopv4f64>;
+ int_x86_fma_vfnmadd_sd>;
+defm VFNMADDPS4 : fma4p<0x78, "vfnmaddps", int_x86_fma_vfnmadd_ps,
+ int_x86_fma_vfnmadd_ps_256, memopv4f32, memopv8f32>;
+defm VFNMADDPD4 : fma4p<0x79, "vfnmaddpd", int_x86_fma_vfnmadd_pd,
+ int_x86_fma_vfnmadd_pd_256, memopv2f64, memopv4f64>;
defm VFNMSUBSS4 : fma4s<0x7E, "vfnmsubss", ssmem, sse_load_f32,
- int_x86_fma4_vfnmsub_ss>;
+ int_x86_fma_vfnmsub_ss>;
defm VFNMSUBSD4 : fma4s<0x7F, "vfnmsubsd", sdmem, sse_load_f64,
- int_x86_fma4_vfnmsub_sd>;
-defm VFNMSUBPS4 : fma4p<0x7C, "vfnmsubps", int_x86_fma4_vfnmsub_ps,
- int_x86_fma4_vfnmsub_ps_256, memopv4f32, memopv8f32>;
-defm VFNMSUBPD4 : fma4p<0x7D, "vfnmsubpd", int_x86_fma4_vfnmsub_pd,
- int_x86_fma4_vfnmsub_pd_256, memopv2f64, memopv4f64>;
-defm VFMADDSUBPS4 : fma4p<0x5C, "vfmaddsubps", int_x86_fma4_vfmaddsub_ps,
- int_x86_fma4_vfmaddsub_ps_256, memopv4f32, memopv8f32>;
-defm VFMADDSUBPD4 : fma4p<0x5D, "vfmaddsubpd", int_x86_fma4_vfmaddsub_pd,
- int_x86_fma4_vfmaddsub_pd_256, memopv2f64, memopv4f64>;
-defm VFMSUBADDPS4 : fma4p<0x5E, "vfmsubaddps", int_x86_fma4_vfmsubadd_ps,
- int_x86_fma4_vfmsubadd_ps_256, memopv4f32, memopv8f32>;
-defm VFMSUBADDPD4 : fma4p<0x5F, "vfmsubaddpd", int_x86_fma4_vfmsubadd_pd,
- int_x86_fma4_vfmsubadd_pd_256, memopv2f64, memopv4f64>;
+ int_x86_fma_vfnmsub_sd>;
+defm VFNMSUBPS4 : fma4p<0x7C, "vfnmsubps", int_x86_fma_vfnmsub_ps,
+ int_x86_fma_vfnmsub_ps_256, memopv4f32, memopv8f32>;
+defm VFNMSUBPD4 : fma4p<0x7D, "vfnmsubpd", int_x86_fma_vfnmsub_pd,
+ int_x86_fma_vfnmsub_pd_256, memopv2f64, memopv4f64>;
+defm VFMADDSUBPS4 : fma4p<0x5C, "vfmaddsubps", int_x86_fma_vfmaddsub_ps,
+ int_x86_fma_vfmaddsub_ps_256, memopv4f32, memopv8f32>;
+defm VFMADDSUBPD4 : fma4p<0x5D, "vfmaddsubpd", int_x86_fma_vfmaddsub_pd,
+ int_x86_fma_vfmaddsub_pd_256, memopv2f64, memopv4f64>;
+defm VFMSUBADDPS4 : fma4p<0x5E, "vfmsubaddps", int_x86_fma_vfmsubadd_ps,
+ int_x86_fma_vfmsubadd_ps_256, memopv4f32, memopv8f32>;
+defm VFMSUBADDPD4 : fma4p<0x5F, "vfmsubaddpd", int_x86_fma_vfmsubadd_pd,
+ int_x86_fma_vfmsubadd_pd_256, memopv2f64, memopv4f64>;
+} // HasFMA4
+
projects / oota-llvm.git / blobdiff