[X86] Improve lowering of v2i64 sign bit tests on pre-sse4.2 targets

Without sse4.2 a v2i64 setlt needs to expand into a pcmpgtd, pcmpeqd, 3 shuffles, and 2 logic ops. But if we're only interested in the sign bit of the i64 elements, we can just use one pcmpgtd and shuffle the odd elements to the even elements.

Differential Revision: https://reviews.llvm.org/D72302

Author: topperc

llvm/lib/Target/X86/X86ISelLowering.cpp

@@ -21584,6 +21584,19 @@ static SDValue LowerVSETCC(SDValue Op, const X86Subtarget &Subtarget,
     if (Opc == X86ISD::PCMPGT && !Subtarget.hasSSE42()) {
       assert(Subtarget.hasSSE2() && "Don't know how to lower!");
 
+      // Special case for sign bit test. We can use a v4i32 PCMPGT and shuffle
+      // the odd elements over the even elements.
+      if (!FlipSigns && !Invert && ISD::isBuildVectorAllZeros(Op0.getNode())) {
+        Op0 = DAG.getConstant(0, dl, MVT::v4i32);
+        Op1 = DAG.getBitcast(MVT::v4i32, Op1);
+
+        SDValue GT = DAG.getNode(X86ISD::PCMPGT, dl, MVT::v4i32, Op0, Op1);
+        static const int MaskHi[] = { 1, 1, 3, 3 };
+        SDValue Result = DAG.getVectorShuffle(MVT::v4i32, dl, GT, GT, MaskHi);
+
+        return DAG.getBitcast(VT, Result);
+      }
+
       // Since SSE has no unsigned integer comparisons, we need to flip the sign
       // bits of the inputs before performing those operations. The lower
       // compare is always unsigned.

llvm/test/CodeGen/X86/bitcast-vector-bool.ll

@@ -429,48 +429,11 @@ define i16 @bitcast_v32i8_to_v2i16(<32 x i8> %a0) nounwind {
 define i4 @bitcast_v8i64_to_v2i4(<8 x i64> %a0) nounwind {
 ; SSE2-SSSE3-LABEL: bitcast_v8i64_to_v2i4:
 ; SSE2-SSSE3:       # %bb.0:
-; SSE2-SSSE3-NEXT:    movdqa {{.*#+}} xmm4 = [2147483648,2147483648]
-; SSE2-SSSE3-NEXT:    pxor %xmm4, %xmm3
-; SSE2-SSSE3-NEXT:    movdqa %xmm4, %xmm5
-; SSE2-SSSE3-NEXT:    pcmpgtd %xmm3, %xmm5
-; SSE2-SSSE3-NEXT:    pshufd {{.*#+}} xmm6 = xmm5[0,0,2,2]
-; SSE2-SSSE3-NEXT:    pcmpeqd %xmm4, %xmm3
-; SSE2-SSSE3-NEXT:    pshufd {{.*#+}} xmm3 = xmm3[1,1,3,3]
-; SSE2-SSSE3-NEXT:    pand %xmm6, %xmm3
-; SSE2-SSSE3-NEXT:    pshufd {{.*#+}} xmm5 = xmm5[1,1,3,3]
-; SSE2-SSSE3-NEXT:    por %xmm3, %xmm5
-; SSE2-SSSE3-NEXT:    pxor %xmm4, %xmm2
-; SSE2-SSSE3-NEXT:    movdqa %xmm4, %xmm3
-; SSE2-SSSE3-NEXT:    pcmpgtd %xmm2, %xmm3
-; SSE2-SSSE3-NEXT:    pshufd {{.*#+}} xmm6 = xmm3[0,0,2,2]
-; SSE2-SSSE3-NEXT:    pcmpeqd %xmm4, %xmm2
-; SSE2-SSSE3-NEXT:    pshufd {{.*#+}} xmm7 = xmm2[1,1,3,3]
-; SSE2-SSSE3-NEXT:    pand %xmm6, %xmm7
-; SSE2-SSSE3-NEXT:    pshufd {{.*#+}} xmm2 = xmm3[1,1,3,3]
-; SSE2-SSSE3-NEXT:    por %xmm7, %xmm2
-; SSE2-SSSE3-NEXT:    packssdw %xmm5, %xmm2
-; SSE2-SSSE3-NEXT:    pxor %xmm4, %xmm1
-; SSE2-SSSE3-NEXT:    movdqa %xmm4, %xmm3
-; SSE2-SSSE3-NEXT:    pcmpgtd %xmm1, %xmm3
-; SSE2-SSSE3-NEXT:    pshufd {{.*#+}} xmm5 = xmm3[0,0,2,2]
-; SSE2-SSSE3-NEXT:    pcmpeqd %xmm4, %xmm1
-; SSE2-SSSE3-NEXT:    pshufd {{.*#+}} xmm1 = xmm1[1,1,3,3]
-; SSE2-SSSE3-NEXT:    pand %xmm5, %xmm1
-; SSE2-SSSE3-NEXT:    pshufd {{.*#+}} xmm3 = xmm3[1,1,3,3]
-; SSE2-SSSE3-NEXT:    por %xmm1, %xmm3
-; SSE2-SSSE3-NEXT:    pxor %xmm4, %xmm0
-; SSE2-SSSE3-NEXT:    movdqa %xmm4, %xmm1
-; SSE2-SSSE3-NEXT:    pcmpgtd %xmm0, %xmm1
-; SSE2-SSSE3-NEXT:    pshufd {{.*#+}} xmm5 = xmm1[0,0,2,2]
-; SSE2-SSSE3-NEXT:    pcmpeqd %xmm4, %xmm0
-; SSE2-SSSE3-NEXT:    pshufd {{.*#+}} xmm0 = xmm0[1,1,3,3]
-; SSE2-SSSE3-NEXT:    pand %xmm5, %xmm0
-; SSE2-SSSE3-NEXT:    pshufd {{.*#+}} xmm1 = xmm1[1,1,3,3]
-; SSE2-SSSE3-NEXT:    por %xmm0, %xmm1
-; SSE2-SSSE3-NEXT:    packssdw %xmm3, %xmm1
-; SSE2-SSSE3-NEXT:    packssdw %xmm2, %xmm1
-; SSE2-SSSE3-NEXT:    packsswb %xmm0, %xmm1
-; SSE2-SSSE3-NEXT:    pmovmskb %xmm1, %eax
+; SSE2-SSSE3-NEXT:    packssdw %xmm3, %xmm2
+; SSE2-SSSE3-NEXT:    packssdw %xmm1, %xmm0
+; SSE2-SSSE3-NEXT:    packssdw %xmm2, %xmm0
+; SSE2-SSSE3-NEXT:    packsswb %xmm0, %xmm0
+; SSE2-SSSE3-NEXT:    pmovmskb %xmm0, %eax
 ; SSE2-SSSE3-NEXT:    movzbl %al, %ecx
 ; SSE2-SSSE3-NEXT:    shrl $4, %ecx
 ; SSE2-SSSE3-NEXT:    movq %rcx, %xmm0

llvm/test/CodeGen/X86/movmsk-cmp.ll

@@ -1015,48 +1015,11 @@ define i1 @allzeros_v4i64_sign(<4 x i64> %arg) {
 define i1 @allones_v8i64_sign(<8 x i64> %arg) {
 ; SSE2-LABEL: allones_v8i64_sign:
 ; SSE2:       # %bb.0:
-; SSE2-NEXT:    movdqa {{.*#+}} xmm4 = [2147483648,2147483648]
-; SSE2-NEXT:    pxor %xmm4, %xmm3
-; SSE2-NEXT:    movdqa %xmm4, %xmm5
-; SSE2-NEXT:    pcmpgtd %xmm3, %xmm5
-; SSE2-NEXT:    pshufd {{.*#+}} xmm6 = xmm5[0,0,2,2]
-; SSE2-NEXT:    pcmpeqd %xmm4, %xmm3
-; SSE2-NEXT:    pshufd {{.*#+}} xmm3 = xmm3[1,1,3,3]
-; SSE2-NEXT:    pand %xmm6, %xmm3
-; SSE2-NEXT:    pshufd {{.*#+}} xmm5 = xmm5[1,1,3,3]
-; SSE2-NEXT:    por %xmm3, %xmm5
-; SSE2-NEXT:    pxor %xmm4, %xmm2
-; SSE2-NEXT:    movdqa %xmm4, %xmm3
-; SSE2-NEXT:    pcmpgtd %xmm2, %xmm3
-; SSE2-NEXT:    pshufd {{.*#+}} xmm6 = xmm3[0,0,2,2]
-; SSE2-NEXT:    pcmpeqd %xmm4, %xmm2
-; SSE2-NEXT:    pshufd {{.*#+}} xmm7 = xmm2[1,1,3,3]
-; SSE2-NEXT:    pand %xmm6, %xmm7
-; SSE2-NEXT:    pshufd {{.*#+}} xmm2 = xmm3[1,1,3,3]
-; SSE2-NEXT:    por %xmm7, %xmm2
-; SSE2-NEXT:    packssdw %xmm5, %xmm2
-; SSE2-NEXT:    pxor %xmm4, %xmm1
-; SSE2-NEXT:    movdqa %xmm4, %xmm3
-; SSE2-NEXT:    pcmpgtd %xmm1, %xmm3
-; SSE2-NEXT:    pshufd {{.*#+}} xmm5 = xmm3[0,0,2,2]
-; SSE2-NEXT:    pcmpeqd %xmm4, %xmm1
-; SSE2-NEXT:    pshufd {{.*#+}} xmm1 = xmm1[1,1,3,3]
-; SSE2-NEXT:    pand %xmm5, %xmm1
-; SSE2-NEXT:    pshufd {{.*#+}} xmm3 = xmm3[1,1,3,3]
-; SSE2-NEXT:    por %xmm1, %xmm3
-; SSE2-NEXT:    pxor %xmm4, %xmm0
-; SSE2-NEXT:    movdqa %xmm4, %xmm1
-; SSE2-NEXT:    pcmpgtd %xmm0, %xmm1
-; SSE2-NEXT:    pshufd {{.*#+}} xmm5 = xmm1[0,0,2,2]
-; SSE2-NEXT:    pcmpeqd %xmm4, %xmm0
-; SSE2-NEXT:    pshufd {{.*#+}} xmm0 = xmm0[1,1,3,3]
-; SSE2-NEXT:    pand %xmm5, %xmm0
-; SSE2-NEXT:    pshufd {{.*#+}} xmm1 = xmm1[1,1,3,3]
-; SSE2-NEXT:    por %xmm0, %xmm1
-; SSE2-NEXT:    packssdw %xmm3, %xmm1
-; SSE2-NEXT:    packssdw %xmm2, %xmm1
-; SSE2-NEXT:    packsswb %xmm0, %xmm1
-; SSE2-NEXT:    pmovmskb %xmm1, %eax
+; SSE2-NEXT:    packssdw %xmm3, %xmm2
+; SSE2-NEXT:    packssdw %xmm1, %xmm0
+; SSE2-NEXT:    packssdw %xmm2, %xmm0
+; SSE2-NEXT:    packsswb %xmm0, %xmm0
+; SSE2-NEXT:    pmovmskb %xmm0, %eax
 ; SSE2-NEXT:    cmpb $-1, %al
 ; SSE2-NEXT:    sete %al
 ; SSE2-NEXT:    retq
@@ -1113,48 +1076,11 @@ define i1 @allones_v8i64_sign(<8 x i64> %arg) {
 define i1 @allzeros_v8i64_sign(<8 x i64> %arg) {
 ; SSE2-LABEL: allzeros_v8i64_sign:
 ; SSE2:       # %bb.0:
-; SSE2-NEXT:    movdqa {{.*#+}} xmm4 = [2147483648,2147483648]
-; SSE2-NEXT:    pxor %xmm4, %xmm3
-; SSE2-NEXT:    movdqa %xmm4, %xmm5
-; SSE2-NEXT:    pcmpgtd %xmm3, %xmm5
-; SSE2-NEXT:    pshufd {{.*#+}} xmm6 = xmm5[0,0,2,2]
-; SSE2-NEXT:    pcmpeqd %xmm4, %xmm3
-; SSE2-NEXT:    pshufd {{.*#+}} xmm3 = xmm3[1,1,3,3]
-; SSE2-NEXT:    pand %xmm6, %xmm3
-; SSE2-NEXT:    pshufd {{.*#+}} xmm5 = xmm5[1,1,3,3]
-; SSE2-NEXT:    por %xmm3, %xmm5
-; SSE2-NEXT:    pxor %xmm4, %xmm2
-; SSE2-NEXT:    movdqa %xmm4, %xmm3
-; SSE2-NEXT:    pcmpgtd %xmm2, %xmm3
-; SSE2-NEXT:    pshufd {{.*#+}} xmm6 = xmm3[0,0,2,2]
-; SSE2-NEXT:    pcmpeqd %xmm4, %xmm2
-; SSE2-NEXT:    pshufd {{.*#+}} xmm7 = xmm2[1,1,3,3]
-; SSE2-NEXT:    pand %xmm6, %xmm7
-; SSE2-NEXT:    pshufd {{.*#+}} xmm2 = xmm3[1,1,3,3]
-; SSE2-NEXT:    por %xmm7, %xmm2
-; SSE2-NEXT:    packssdw %xmm5, %xmm2
-; SSE2-NEXT:    pxor %xmm4, %xmm1
-; SSE2-NEXT:    movdqa %xmm4, %xmm3
-; SSE2-NEXT:    pcmpgtd %xmm1, %xmm3
-; SSE2-NEXT:    pshufd {{.*#+}} xmm5 = xmm3[0,0,2,2]
-; SSE2-NEXT:    pcmpeqd %xmm4, %xmm1
-; SSE2-NEXT:    pshufd {{.*#+}} xmm1 = xmm1[1,1,3,3]
-; SSE2-NEXT:    pand %xmm5, %xmm1
-; SSE2-NEXT:    pshufd {{.*#+}} xmm3 = xmm3[1,1,3,3]
-; SSE2-NEXT:    por %xmm1, %xmm3
-; SSE2-NEXT:    pxor %xmm4, %xmm0
-; SSE2-NEXT:    movdqa %xmm4, %xmm1
-; SSE2-NEXT:    pcmpgtd %xmm0, %xmm1
-; SSE2-NEXT:    pshufd {{.*#+}} xmm5 = xmm1[0,0,2,2]
-; SSE2-NEXT:    pcmpeqd %xmm4, %xmm0
-; SSE2-NEXT:    pshufd {{.*#+}} xmm0 = xmm0[1,1,3,3]
-; SSE2-NEXT:    pand %xmm5, %xmm0
-; SSE2-NEXT:    pshufd {{.*#+}} xmm1 = xmm1[1,1,3,3]
-; SSE2-NEXT:    por %xmm0, %xmm1
-; SSE2-NEXT:    packssdw %xmm3, %xmm1
-; SSE2-NEXT:    packssdw %xmm2, %xmm1
-; SSE2-NEXT:    packsswb %xmm0, %xmm1
-; SSE2-NEXT:    pmovmskb %xmm1, %eax
+; SSE2-NEXT:    packssdw %xmm3, %xmm2
+; SSE2-NEXT:    packssdw %xmm1, %xmm0
+; SSE2-NEXT:    packssdw %xmm2, %xmm0
+; SSE2-NEXT:    packsswb %xmm0, %xmm0
+; SSE2-NEXT:    pmovmskb %xmm0, %eax
 ; SSE2-NEXT:    testb %al, %al
 ; SSE2-NEXT:    sete %al
 ; SSE2-NEXT:    retq