[SLP][NFC]Add a test with incomplete insertion mask, NFC

Author: alexey-bataev

llvm/test/Transforms/SLPVectorizer/X86/insert-subvector.ll

@@ -0,0 +1,40 @@
+; NOTE: Assertions have been autogenerated by utils/update_test_checks.py UTC_ARGS: --version 3
+; RUN: opt < %s -passes=slp-vectorizer -S -mtriple=x86_64-unknown-linux-gnu -mcpu=skylake-avx512 | FileCheck %s
+
+define <8 x float> @test(ptr %x, float %v, float %a) {
+; CHECK-LABEL: define <8 x float> @test(
+; CHECK-SAME: ptr [[X:%.*]], float [[V:%.*]], float [[A:%.*]]) #[[ATTR0:[0-9]+]] {
+; CHECK-NEXT:    [[TMP2:%.*]] = insertelement <4 x float> poison, float [[A]], i32 0
+; CHECK-NEXT:    [[TMP3:%.*]] = shufflevector <4 x float> [[TMP2]], <4 x float> poison, <4 x i32> zeroinitializer
+; CHECK-NEXT:    [[TMP4:%.*]] = insertelement <4 x float> poison, float [[V]], i32 0
+; CHECK-NEXT:    [[TMP7:%.*]] = shufflevector <4 x float> [[TMP4]], <4 x float> poison, <4 x i32> zeroinitializer
+; CHECK-NEXT:    [[TMP8:%.*]] = fadd <4 x float> [[TMP3]], [[TMP7]]
+; CHECK-NEXT:    [[TMP6:%.*]] = load <2 x float>, ptr [[X]], align 4
+; CHECK-NEXT:    [[TMP11:%.*]] = call <4 x float> @llvm.vector.insert.v4f32.v2f32(<4 x float> [[TMP7]], <2 x float> [[TMP6]], i64 0)
+; CHECK-NEXT:    [[TMP12:%.*]] = fadd <4 x float> [[TMP3]], [[TMP11]]
+; CHECK-NEXT:    [[TMP9:%.*]] = shufflevector <4 x float> [[TMP8]], <4 x float> undef, <8 x i32> <i32 0, i32 1, i32 2, i32 3, i32 4, i32 5, i32 6, i32 7>
+; CHECK-NEXT:    [[TMP10:%.*]] = shufflevector <4 x float> [[TMP12]], <4 x float> poison, <8 x i32> <i32 0, i32 1, i32 2, i32 3, i32 poison, i32 poison, i32 poison, i32 poison>
+; CHECK-NEXT:    [[I71:%.*]] = shufflevector <8 x float> [[TMP9]], <8 x float> [[TMP10]], <8 x i32> <i32 8, i32 9, i32 10, i32 11, i32 4, i32 5, i32 6, i32 7>
+; CHECK-NEXT:    ret <8 x float> [[I71]]
+;
+  %gep1 = getelementptr inbounds <4 x float>, ptr %x, i64 0, i64 1
+  %x0 = load float, ptr %x, align 4
+  %x1 = load float, ptr %gep1, align 4
+  %add1 = fadd float %a, %v
+  %add2 = fadd float %a, %v
+  %add3 = fadd float %a, %v
+  %add4 = fadd float %a, %v
+  %add5 = fadd float %a, %x0
+  %add6 = fadd float %a, %x1
+  %add7 = fadd float %a, %v
+  %add8 = fadd float %a, %v
+  %i0 = insertelement <8 x float> undef, float %add1, i32 0
+  %i1 = insertelement <8 x float> %i0, float %add2, i32 1
+  %i2 = insertelement <8 x float> %i1, float %add3, i32 2
+  %i3 = insertelement <8 x float> %i2, float %add4, i32 3
+  %i4 = insertelement <8 x float> %i3, float %add5, i32 0
+  %i5 = insertelement <8 x float> %i4, float %add6, i32 1
+  %i6 = insertelement <8 x float> %i5, float %add7, i32 2
+  %i7 = insertelement <8 x float> %i6, float %add8, i32 3
+  ret <8 x float> %i7
+}