[X86] combineConcatVectorOps - add handling for ISD::SMIN/SMAX/UMIN/UMAX vector ops (llvm#172667)

Author: RKSimon

llvm/lib/Target/X86/X86ISelLowering.cpp

@@ -59632,6 +59632,10 @@ static SDValue combineConcatVectorOps(const SDLoc &DL, MVT VT,
     case ISD::ADD:
     case ISD::SUB:
     case ISD::MUL:
+    case ISD::SMIN:
+    case ISD::SMAX:
+    case ISD::UMIN:
+    case ISD::UMAX:
       // TODO: Add more integer binops?
       if (!IsSplat && ((VT.is256BitVector() && Subtarget.hasInt256()) ||
                        (VT.is512BitVector() && Subtarget.useAVX512Regs() &&

llvm/test/CodeGen/X86/combine-icmp.ll

@@ -532,18 +532,16 @@ define i64 @concat_icmp_v64i8_v16i8(<16 x i8> %a0, <16 x i8> %a1, <16 x i8> %a2,
 ;
 ; AVX2-LABEL: concat_icmp_v64i8_v16i8:
 ; AVX2:       # %bb.0:
-; AVX2-NEXT:    vpbroadcastd {{.*#+}} xmm4 = [16,16,16,16,16,16,16,16,16,16,16,16,16,16,16,16]
-; AVX2-NEXT:    vpmaxub %xmm4, %xmm0, %xmm5
-; AVX2-NEXT:    vpcmpeqb %xmm5, %xmm0, %xmm0
-; AVX2-NEXT:    vpmaxub %xmm4, %xmm1, %xmm5
-; AVX2-NEXT:    vpcmpeqb %xmm5, %xmm1, %xmm1
-; AVX2-NEXT:    vpmaxub %xmm4, %xmm2, %xmm5
-; AVX2-NEXT:    vpcmpeqb %xmm5, %xmm2, %xmm2
-; AVX2-NEXT:    vpmaxub %xmm4, %xmm3, %xmm4
-; AVX2-NEXT:    vpcmpeqb %xmm4, %xmm3, %xmm3
+; AVX2-NEXT:    # kill: def $xmm2 killed $xmm2 def $ymm2
+; AVX2-NEXT:    # kill: def $xmm0 killed $xmm0 def $ymm0
 ; AVX2-NEXT:    vinserti128 $1, %xmm1, %ymm0, %ymm0
+; AVX2-NEXT:    vpbroadcastd {{.*#+}} ymm1 = [16,16,16,16,16,16,16,16,16,16,16,16,16,16,16,16,16,16,16,16,16,16,16,16,16,16,16,16,16,16,16,16]
+; AVX2-NEXT:    vpmaxub %ymm1, %ymm0, %ymm4
+; AVX2-NEXT:    vpcmpeqb %ymm4, %ymm0, %ymm0
 ; AVX2-NEXT:    vpmovmskb %ymm0, %ecx
 ; AVX2-NEXT:    vinserti128 $1, %xmm3, %ymm2, %ymm0
+; AVX2-NEXT:    vpmaxub %ymm1, %ymm0, %ymm1
+; AVX2-NEXT:    vpcmpeqb %ymm1, %ymm0, %ymm0
 ; AVX2-NEXT:    vpmovmskb %ymm0, %eax
 ; AVX2-NEXT:    shlq $32, %rax
 ; AVX2-NEXT:    orq %rcx, %rax