summaryrefslogtreecommitdiffstats
path: root/llvm/test
diff options
context:
space:
mode:
Diffstat (limited to 'llvm/test')
-rw-r--r--llvm/test/Transforms/InstCombine/x86-sse.ll308
-rw-r--r--llvm/test/Transforms/InstCombine/x86-sse2.ll206
2 files changed, 511 insertions, 3 deletions
diff --git a/llvm/test/Transforms/InstCombine/x86-sse.ll b/llvm/test/Transforms/InstCombine/x86-sse.ll
index f0cdf0b1165..d6b4796ed91 100644
--- a/llvm/test/Transforms/InstCombine/x86-sse.ll
+++ b/llvm/test/Transforms/InstCombine/x86-sse.ll
@@ -207,9 +207,297 @@ define float @test_cmp_ss_0(float %a, float %b) {
ret float %r
}
-declare <4 x float> @llvm.x86.sse.rcp.ss(<4 x float>) nounwind readnone
-declare <4 x float> @llvm.x86.sse.sqrt.ss(<4 x float>) nounwind readnone
-declare <4 x float> @llvm.x86.sse.rsqrt.ss(<4 x float>) nounwind readnone
+define i32 @test_comieq_ss_0(float %a, float %b) {
+; CHECK-LABEL: @test_comieq_ss_0
+; CHECK-NEXT: %1 = insertelement <4 x float> undef, float %a, i32 0
+; CHECK-NEXT: %2 = insertelement <4 x float> %1, float 1.000000e+00, i32 1
+; CHECK-NEXT: %3 = insertelement <4 x float> %2, float 2.000000e+00, i32 2
+; CHECK-NEXT: %4 = insertelement <4 x float> %3, float 3.000000e+00, i32 3
+; CHECK-NEXT: %5 = insertelement <4 x float> undef, float %b, i32 0
+; CHECK-NEXT: %6 = insertelement <4 x float> %5, float 4.000000e+00, i32 1
+; CHECK-NEXT: %7 = insertelement <4 x float> %6, float 5.000000e+00, i32 2
+; CHECK-NEXT: %8 = insertelement <4 x float> %7, float 6.000000e+00, i32 3
+; CHECK-NEXT: %9 = tail call i32 @llvm.x86.sse.comieq.ss(<4 x float> %4, <4 x float> %8)
+; CHECK-NEXT: ret i32 %9
+ %1 = insertelement <4 x float> undef, float %a, i32 0
+ %2 = insertelement <4 x float> %1, float 1.000000e+00, i32 1
+ %3 = insertelement <4 x float> %2, float 2.000000e+00, i32 2
+ %4 = insertelement <4 x float> %3, float 3.000000e+00, i32 3
+ %5 = insertelement <4 x float> undef, float %b, i32 0
+ %6 = insertelement <4 x float> %5, float 4.000000e+00, i32 1
+ %7 = insertelement <4 x float> %6, float 5.000000e+00, i32 2
+ %8 = insertelement <4 x float> %7, float 6.000000e+00, i32 3
+ %9 = tail call i32 @llvm.x86.sse.comieq.ss(<4 x float> %4, <4 x float> %8)
+ ret i32 %9
+}
+
+define i32 @test_comige_ss_0(float %a, float %b) {
+; CHECK-LABEL: @test_comige_ss_0
+; CHECK-NEXT: %1 = insertelement <4 x float> undef, float %a, i32 0
+; CHECK-NEXT: %2 = insertelement <4 x float> %1, float 1.000000e+00, i32 1
+; CHECK-NEXT: %3 = insertelement <4 x float> %2, float 2.000000e+00, i32 2
+; CHECK-NEXT: %4 = insertelement <4 x float> %3, float 3.000000e+00, i32 3
+; CHECK-NEXT: %5 = insertelement <4 x float> undef, float %b, i32 0
+; CHECK-NEXT: %6 = insertelement <4 x float> %5, float 4.000000e+00, i32 1
+; CHECK-NEXT: %7 = insertelement <4 x float> %6, float 5.000000e+00, i32 2
+; CHECK-NEXT: %8 = insertelement <4 x float> %7, float 6.000000e+00, i32 3
+; CHECK-NEXT: %9 = tail call i32 @llvm.x86.sse.comige.ss(<4 x float> %4, <4 x float> %8)
+; CHECK-NEXT: ret i32 %9
+ %1 = insertelement <4 x float> undef, float %a, i32 0
+ %2 = insertelement <4 x float> %1, float 1.000000e+00, i32 1
+ %3 = insertelement <4 x float> %2, float 2.000000e+00, i32 2
+ %4 = insertelement <4 x float> %3, float 3.000000e+00, i32 3
+ %5 = insertelement <4 x float> undef, float %b, i32 0
+ %6 = insertelement <4 x float> %5, float 4.000000e+00, i32 1
+ %7 = insertelement <4 x float> %6, float 5.000000e+00, i32 2
+ %8 = insertelement <4 x float> %7, float 6.000000e+00, i32 3
+ %9 = tail call i32 @llvm.x86.sse.comige.ss(<4 x float> %4, <4 x float> %8)
+ ret i32 %9
+}
+
+define i32 @test_comigt_ss_0(float %a, float %b) {
+; CHECK-LABEL: @test_comigt_ss_0
+; CHECK-NEXT: %1 = insertelement <4 x float> undef, float %a, i32 0
+; CHECK-NEXT: %2 = insertelement <4 x float> %1, float 1.000000e+00, i32 1
+; CHECK-NEXT: %3 = insertelement <4 x float> %2, float 2.000000e+00, i32 2
+; CHECK-NEXT: %4 = insertelement <4 x float> %3, float 3.000000e+00, i32 3
+; CHECK-NEXT: %5 = insertelement <4 x float> undef, float %b, i32 0
+; CHECK-NEXT: %6 = insertelement <4 x float> %5, float 4.000000e+00, i32 1
+; CHECK-NEXT: %7 = insertelement <4 x float> %6, float 5.000000e+00, i32 2
+; CHECK-NEXT: %8 = insertelement <4 x float> %7, float 6.000000e+00, i32 3
+; CHECK-NEXT: %9 = tail call i32 @llvm.x86.sse.comigt.ss(<4 x float> %4, <4 x float> %8)
+; CHECK-NEXT: ret i32 %9
+ %1 = insertelement <4 x float> undef, float %a, i32 0
+ %2 = insertelement <4 x float> %1, float 1.000000e+00, i32 1
+ %3 = insertelement <4 x float> %2, float 2.000000e+00, i32 2
+ %4 = insertelement <4 x float> %3, float 3.000000e+00, i32 3
+ %5 = insertelement <4 x float> undef, float %b, i32 0
+ %6 = insertelement <4 x float> %5, float 4.000000e+00, i32 1
+ %7 = insertelement <4 x float> %6, float 5.000000e+00, i32 2
+ %8 = insertelement <4 x float> %7, float 6.000000e+00, i32 3
+ %9 = tail call i32 @llvm.x86.sse.comigt.ss(<4 x float> %4, <4 x float> %8)
+ ret i32 %9
+}
+
+define i32 @test_comile_ss_0(float %a, float %b) {
+; CHECK-LABEL: @test_comile_ss_0
+; CHECK-NEXT: %1 = insertelement <4 x float> undef, float %a, i32 0
+; CHECK-NEXT: %2 = insertelement <4 x float> %1, float 1.000000e+00, i32 1
+; CHECK-NEXT: %3 = insertelement <4 x float> %2, float 2.000000e+00, i32 2
+; CHECK-NEXT: %4 = insertelement <4 x float> %3, float 3.000000e+00, i32 3
+; CHECK-NEXT: %5 = insertelement <4 x float> undef, float %b, i32 0
+; CHECK-NEXT: %6 = insertelement <4 x float> %5, float 4.000000e+00, i32 1
+; CHECK-NEXT: %7 = insertelement <4 x float> %6, float 5.000000e+00, i32 2
+; CHECK-NEXT: %8 = insertelement <4 x float> %7, float 6.000000e+00, i32 3
+; CHECK-NEXT: %9 = tail call i32 @llvm.x86.sse.comile.ss(<4 x float> %4, <4 x float> %8)
+; CHECK-NEXT: ret i32 %9
+ %1 = insertelement <4 x float> undef, float %a, i32 0
+ %2 = insertelement <4 x float> %1, float 1.000000e+00, i32 1
+ %3 = insertelement <4 x float> %2, float 2.000000e+00, i32 2
+ %4 = insertelement <4 x float> %3, float 3.000000e+00, i32 3
+ %5 = insertelement <4 x float> undef, float %b, i32 0
+ %6 = insertelement <4 x float> %5, float 4.000000e+00, i32 1
+ %7 = insertelement <4 x float> %6, float 5.000000e+00, i32 2
+ %8 = insertelement <4 x float> %7, float 6.000000e+00, i32 3
+ %9 = tail call i32 @llvm.x86.sse.comile.ss(<4 x float> %4, <4 x float> %8)
+ ret i32 %9
+}
+
+define i32 @test_comilt_ss_0(float %a, float %b) {
+; CHECK-LABEL: @test_comilt_ss_0
+; CHECK-NEXT: %1 = insertelement <4 x float> undef, float %a, i32 0
+; CHECK-NEXT: %2 = insertelement <4 x float> %1, float 1.000000e+00, i32 1
+; CHECK-NEXT: %3 = insertelement <4 x float> %2, float 2.000000e+00, i32 2
+; CHECK-NEXT: %4 = insertelement <4 x float> %3, float 3.000000e+00, i32 3
+; CHECK-NEXT: %5 = insertelement <4 x float> undef, float %b, i32 0
+; CHECK-NEXT: %6 = insertelement <4 x float> %5, float 4.000000e+00, i32 1
+; CHECK-NEXT: %7 = insertelement <4 x float> %6, float 5.000000e+00, i32 2
+; CHECK-NEXT: %8 = insertelement <4 x float> %7, float 6.000000e+00, i32 3
+; CHECK-NEXT: %9 = tail call i32 @llvm.x86.sse.comilt.ss(<4 x float> %4, <4 x float> %8)
+; CHECK-NEXT: ret i32 %9
+ %1 = insertelement <4 x float> undef, float %a, i32 0
+ %2 = insertelement <4 x float> %1, float 1.000000e+00, i32 1
+ %3 = insertelement <4 x float> %2, float 2.000000e+00, i32 2
+ %4 = insertelement <4 x float> %3, float 3.000000e+00, i32 3
+ %5 = insertelement <4 x float> undef, float %b, i32 0
+ %6 = insertelement <4 x float> %5, float 4.000000e+00, i32 1
+ %7 = insertelement <4 x float> %6, float 5.000000e+00, i32 2
+ %8 = insertelement <4 x float> %7, float 6.000000e+00, i32 3
+ %9 = tail call i32 @llvm.x86.sse.comilt.ss(<4 x float> %4, <4 x float> %8)
+ ret i32 %9
+}
+
+define i32 @test_comineq_ss_0(float %a, float %b) {
+; CHECK-LABEL: @test_comineq_ss_0
+; CHECK-NEXT: %1 = insertelement <4 x float> undef, float %a, i32 0
+; CHECK-NEXT: %2 = insertelement <4 x float> %1, float 1.000000e+00, i32 1
+; CHECK-NEXT: %3 = insertelement <4 x float> %2, float 2.000000e+00, i32 2
+; CHECK-NEXT: %4 = insertelement <4 x float> %3, float 3.000000e+00, i32 3
+; CHECK-NEXT: %5 = insertelement <4 x float> undef, float %b, i32 0
+; CHECK-NEXT: %6 = insertelement <4 x float> %5, float 4.000000e+00, i32 1
+; CHECK-NEXT: %7 = insertelement <4 x float> %6, float 5.000000e+00, i32 2
+; CHECK-NEXT: %8 = insertelement <4 x float> %7, float 6.000000e+00, i32 3
+; CHECK-NEXT: %9 = tail call i32 @llvm.x86.sse.comineq.ss(<4 x float> %4, <4 x float> %8)
+; CHECK-NEXT: ret i32 %9
+ %1 = insertelement <4 x float> undef, float %a, i32 0
+ %2 = insertelement <4 x float> %1, float 1.000000e+00, i32 1
+ %3 = insertelement <4 x float> %2, float 2.000000e+00, i32 2
+ %4 = insertelement <4 x float> %3, float 3.000000e+00, i32 3
+ %5 = insertelement <4 x float> undef, float %b, i32 0
+ %6 = insertelement <4 x float> %5, float 4.000000e+00, i32 1
+ %7 = insertelement <4 x float> %6, float 5.000000e+00, i32 2
+ %8 = insertelement <4 x float> %7, float 6.000000e+00, i32 3
+ %9 = tail call i32 @llvm.x86.sse.comineq.ss(<4 x float> %4, <4 x float> %8)
+ ret i32 %9
+}
+
+define i32 @test_ucomieq_ss_0(float %a, float %b) {
+; CHECK-LABEL: @test_ucomieq_ss_0
+; CHECK-NEXT: %1 = insertelement <4 x float> undef, float %a, i32 0
+; CHECK-NEXT: %2 = insertelement <4 x float> %1, float 1.000000e+00, i32 1
+; CHECK-NEXT: %3 = insertelement <4 x float> %2, float 2.000000e+00, i32 2
+; CHECK-NEXT: %4 = insertelement <4 x float> %3, float 3.000000e+00, i32 3
+; CHECK-NEXT: %5 = insertelement <4 x float> undef, float %b, i32 0
+; CHECK-NEXT: %6 = insertelement <4 x float> %5, float 4.000000e+00, i32 1
+; CHECK-NEXT: %7 = insertelement <4 x float> %6, float 5.000000e+00, i32 2
+; CHECK-NEXT: %8 = insertelement <4 x float> %7, float 6.000000e+00, i32 3
+; CHECK-NEXT: %9 = tail call i32 @llvm.x86.sse.ucomieq.ss(<4 x float> %4, <4 x float> %8)
+; CHECK-NEXT: ret i32 %9
+ %1 = insertelement <4 x float> undef, float %a, i32 0
+ %2 = insertelement <4 x float> %1, float 1.000000e+00, i32 1
+ %3 = insertelement <4 x float> %2, float 2.000000e+00, i32 2
+ %4 = insertelement <4 x float> %3, float 3.000000e+00, i32 3
+ %5 = insertelement <4 x float> undef, float %b, i32 0
+ %6 = insertelement <4 x float> %5, float 4.000000e+00, i32 1
+ %7 = insertelement <4 x float> %6, float 5.000000e+00, i32 2
+ %8 = insertelement <4 x float> %7, float 6.000000e+00, i32 3
+ %9 = tail call i32 @llvm.x86.sse.ucomieq.ss(<4 x float> %4, <4 x float> %8)
+ ret i32 %9
+}
+
+define i32 @test_ucomige_ss_0(float %a, float %b) {
+; CHECK-LABEL: @test_ucomige_ss_0
+; CHECK-NEXT: %1 = insertelement <4 x float> undef, float %a, i32 0
+; CHECK-NEXT: %2 = insertelement <4 x float> %1, float 1.000000e+00, i32 1
+; CHECK-NEXT: %3 = insertelement <4 x float> %2, float 2.000000e+00, i32 2
+; CHECK-NEXT: %4 = insertelement <4 x float> %3, float 3.000000e+00, i32 3
+; CHECK-NEXT: %5 = insertelement <4 x float> undef, float %b, i32 0
+; CHECK-NEXT: %6 = insertelement <4 x float> %5, float 4.000000e+00, i32 1
+; CHECK-NEXT: %7 = insertelement <4 x float> %6, float 5.000000e+00, i32 2
+; CHECK-NEXT: %8 = insertelement <4 x float> %7, float 6.000000e+00, i32 3
+; CHECK-NEXT: %9 = tail call i32 @llvm.x86.sse.ucomige.ss(<4 x float> %4, <4 x float> %8)
+; CHECK-NEXT: ret i32 %9
+ %1 = insertelement <4 x float> undef, float %a, i32 0
+ %2 = insertelement <4 x float> %1, float 1.000000e+00, i32 1
+ %3 = insertelement <4 x float> %2, float 2.000000e+00, i32 2
+ %4 = insertelement <4 x float> %3, float 3.000000e+00, i32 3
+ %5 = insertelement <4 x float> undef, float %b, i32 0
+ %6 = insertelement <4 x float> %5, float 4.000000e+00, i32 1
+ %7 = insertelement <4 x float> %6, float 5.000000e+00, i32 2
+ %8 = insertelement <4 x float> %7, float 6.000000e+00, i32 3
+ %9 = tail call i32 @llvm.x86.sse.ucomige.ss(<4 x float> %4, <4 x float> %8)
+ ret i32 %9
+}
+
+define i32 @test_ucomigt_ss_0(float %a, float %b) {
+; CHECK-LABEL: @test_ucomigt_ss_0
+; CHECK-NEXT: %1 = insertelement <4 x float> undef, float %a, i32 0
+; CHECK-NEXT: %2 = insertelement <4 x float> %1, float 1.000000e+00, i32 1
+; CHECK-NEXT: %3 = insertelement <4 x float> %2, float 2.000000e+00, i32 2
+; CHECK-NEXT: %4 = insertelement <4 x float> %3, float 3.000000e+00, i32 3
+; CHECK-NEXT: %5 = insertelement <4 x float> undef, float %b, i32 0
+; CHECK-NEXT: %6 = insertelement <4 x float> %5, float 4.000000e+00, i32 1
+; CHECK-NEXT: %7 = insertelement <4 x float> %6, float 5.000000e+00, i32 2
+; CHECK-NEXT: %8 = insertelement <4 x float> %7, float 6.000000e+00, i32 3
+; CHECK-NEXT: %9 = tail call i32 @llvm.x86.sse.ucomigt.ss(<4 x float> %4, <4 x float> %8)
+; CHECK-NEXT: ret i32 %9
+ %1 = insertelement <4 x float> undef, float %a, i32 0
+ %2 = insertelement <4 x float> %1, float 1.000000e+00, i32 1
+ %3 = insertelement <4 x float> %2, float 2.000000e+00, i32 2
+ %4 = insertelement <4 x float> %3, float 3.000000e+00, i32 3
+ %5 = insertelement <4 x float> undef, float %b, i32 0
+ %6 = insertelement <4 x float> %5, float 4.000000e+00, i32 1
+ %7 = insertelement <4 x float> %6, float 5.000000e+00, i32 2
+ %8 = insertelement <4 x float> %7, float 6.000000e+00, i32 3
+ %9 = tail call i32 @llvm.x86.sse.ucomigt.ss(<4 x float> %4, <4 x float> %8)
+ ret i32 %9
+}
+
+define i32 @test_ucomile_ss_0(float %a, float %b) {
+; CHECK-LABEL: @test_ucomile_ss_0
+; CHECK-NEXT: %1 = insertelement <4 x float> undef, float %a, i32 0
+; CHECK-NEXT: %2 = insertelement <4 x float> %1, float 1.000000e+00, i32 1
+; CHECK-NEXT: %3 = insertelement <4 x float> %2, float 2.000000e+00, i32 2
+; CHECK-NEXT: %4 = insertelement <4 x float> %3, float 3.000000e+00, i32 3
+; CHECK-NEXT: %5 = insertelement <4 x float> undef, float %b, i32 0
+; CHECK-NEXT: %6 = insertelement <4 x float> %5, float 4.000000e+00, i32 1
+; CHECK-NEXT: %7 = insertelement <4 x float> %6, float 5.000000e+00, i32 2
+; CHECK-NEXT: %8 = insertelement <4 x float> %7, float 6.000000e+00, i32 3
+; CHECK-NEXT: %9 = tail call i32 @llvm.x86.sse.ucomile.ss(<4 x float> %4, <4 x float> %8)
+; CHECK-NEXT: ret i32 %9
+ %1 = insertelement <4 x float> undef, float %a, i32 0
+ %2 = insertelement <4 x float> %1, float 1.000000e+00, i32 1
+ %3 = insertelement <4 x float> %2, float 2.000000e+00, i32 2
+ %4 = insertelement <4 x float> %3, float 3.000000e+00, i32 3
+ %5 = insertelement <4 x float> undef, float %b, i32 0
+ %6 = insertelement <4 x float> %5, float 4.000000e+00, i32 1
+ %7 = insertelement <4 x float> %6, float 5.000000e+00, i32 2
+ %8 = insertelement <4 x float> %7, float 6.000000e+00, i32 3
+ %9 = tail call i32 @llvm.x86.sse.ucomile.ss(<4 x float> %4, <4 x float> %8)
+ ret i32 %9
+}
+
+define i32 @test_ucomilt_ss_0(float %a, float %b) {
+; CHECK-LABEL: @test_ucomilt_ss_0
+; CHECK-NEXT: %1 = insertelement <4 x float> undef, float %a, i32 0
+; CHECK-NEXT: %2 = insertelement <4 x float> %1, float 1.000000e+00, i32 1
+; CHECK-NEXT: %3 = insertelement <4 x float> %2, float 2.000000e+00, i32 2
+; CHECK-NEXT: %4 = insertelement <4 x float> %3, float 3.000000e+00, i32 3
+; CHECK-NEXT: %5 = insertelement <4 x float> undef, float %b, i32 0
+; CHECK-NEXT: %6 = insertelement <4 x float> %5, float 4.000000e+00, i32 1
+; CHECK-NEXT: %7 = insertelement <4 x float> %6, float 5.000000e+00, i32 2
+; CHECK-NEXT: %8 = insertelement <4 x float> %7, float 6.000000e+00, i32 3
+; CHECK-NEXT: %9 = tail call i32 @llvm.x86.sse.ucomilt.ss(<4 x float> %4, <4 x float> %8)
+; CHECK-NEXT: ret i32 %9
+ %1 = insertelement <4 x float> undef, float %a, i32 0
+ %2 = insertelement <4 x float> %1, float 1.000000e+00, i32 1
+ %3 = insertelement <4 x float> %2, float 2.000000e+00, i32 2
+ %4 = insertelement <4 x float> %3, float 3.000000e+00, i32 3
+ %5 = insertelement <4 x float> undef, float %b, i32 0
+ %6 = insertelement <4 x float> %5, float 4.000000e+00, i32 1
+ %7 = insertelement <4 x float> %6, float 5.000000e+00, i32 2
+ %8 = insertelement <4 x float> %7, float 6.000000e+00, i32 3
+ %9 = tail call i32 @llvm.x86.sse.ucomilt.ss(<4 x float> %4, <4 x float> %8)
+ ret i32 %9
+}
+
+define i32 @test_ucomineq_ss_0(float %a, float %b) {
+; CHECK-LABEL: @test_ucomineq_ss_0
+; CHECK-NEXT: %1 = insertelement <4 x float> undef, float %a, i32 0
+; CHECK-NEXT: %2 = insertelement <4 x float> %1, float 1.000000e+00, i32 1
+; CHECK-NEXT: %3 = insertelement <4 x float> %2, float 2.000000e+00, i32 2
+; CHECK-NEXT: %4 = insertelement <4 x float> %3, float 3.000000e+00, i32 3
+; CHECK-NEXT: %5 = insertelement <4 x float> undef, float %b, i32 0
+; CHECK-NEXT: %6 = insertelement <4 x float> %5, float 4.000000e+00, i32 1
+; CHECK-NEXT: %7 = insertelement <4 x float> %6, float 5.000000e+00, i32 2
+; CHECK-NEXT: %8 = insertelement <4 x float> %7, float 6.000000e+00, i32 3
+; CHECK-NEXT: %9 = tail call i32 @llvm.x86.sse.ucomineq.ss(<4 x float> %4, <4 x float> %8)
+; CHECK-NEXT: ret i32 %9
+ %1 = insertelement <4 x float> undef, float %a, i32 0
+ %2 = insertelement <4 x float> %1, float 1.000000e+00, i32 1
+ %3 = insertelement <4 x float> %2, float 2.000000e+00, i32 2
+ %4 = insertelement <4 x float> %3, float 3.000000e+00, i32 3
+ %5 = insertelement <4 x float> undef, float %b, i32 0
+ %6 = insertelement <4 x float> %5, float 4.000000e+00, i32 1
+ %7 = insertelement <4 x float> %6, float 5.000000e+00, i32 2
+ %8 = insertelement <4 x float> %7, float 6.000000e+00, i32 3
+ %9 = tail call i32 @llvm.x86.sse.ucomineq.ss(<4 x float> %4, <4 x float> %8)
+ ret i32 %9
+}
+
+declare <4 x float> @llvm.x86.sse.rcp.ss(<4 x float>)
+declare <4 x float> @llvm.x86.sse.sqrt.ss(<4 x float>)
+declare <4 x float> @llvm.x86.sse.rsqrt.ss(<4 x float>)
declare <4 x float> @llvm.x86.sse.add.ss(<4 x float>, <4 x float>)
declare <4 x float> @llvm.x86.sse.sub.ss(<4 x float>, <4 x float>)
@@ -218,3 +506,17 @@ declare <4 x float> @llvm.x86.sse.div.ss(<4 x float>, <4 x float>)
declare <4 x float> @llvm.x86.sse.min.ss(<4 x float>, <4 x float>)
declare <4 x float> @llvm.x86.sse.max.ss(<4 x float>, <4 x float>)
declare <4 x float> @llvm.x86.sse.cmp.ss(<4 x float>, <4 x float>, i8)
+
+declare i32 @llvm.x86.sse.comieq.ss(<4 x float>, <4 x float>)
+declare i32 @llvm.x86.sse.comige.ss(<4 x float>, <4 x float>)
+declare i32 @llvm.x86.sse.comigt.ss(<4 x float>, <4 x float>)
+declare i32 @llvm.x86.sse.comile.ss(<4 x float>, <4 x float>)
+declare i32 @llvm.x86.sse.comilt.ss(<4 x float>, <4 x float>)
+declare i32 @llvm.x86.sse.comineq.ss(<4 x float>, <4 x float>)
+
+declare i32 @llvm.x86.sse.ucomieq.ss(<4 x float>, <4 x float>)
+declare i32 @llvm.x86.sse.ucomige.ss(<4 x float>, <4 x float>)
+declare i32 @llvm.x86.sse.ucomigt.ss(<4 x float>, <4 x float>)
+declare i32 @llvm.x86.sse.ucomile.ss(<4 x float>, <4 x float>)
+declare i32 @llvm.x86.sse.ucomilt.ss(<4 x float>, <4 x float>)
+declare i32 @llvm.x86.sse.ucomineq.ss(<4 x float>, <4 x float>)
diff --git a/llvm/test/Transforms/InstCombine/x86-sse2.ll b/llvm/test/Transforms/InstCombine/x86-sse2.ll
index 1b02a7c3216..0fb6af48e75 100644
--- a/llvm/test/Transforms/InstCombine/x86-sse2.ll
+++ b/llvm/test/Transforms/InstCombine/x86-sse2.ll
@@ -127,6 +127,198 @@ define double @test_cmp_sd_0(double %a, double %b) {
ret double %6
}
+define i32 @test_comieq_sd_0(double %a, double %b) {
+; CHECK-LABEL: @test_comieq_sd_0
+; CHECK-NEXT: %1 = insertelement <2 x double> undef, double %a, i32 0
+; CHECK-NEXT: %2 = insertelement <2 x double> %1, double 1.000000e+00, i32 1
+; CHECK-NEXT: %3 = insertelement <2 x double> undef, double %b, i32 0
+; CHECK-NEXT: %4 = insertelement <2 x double> %3, double 2.000000e+00, i32 1
+; CHECK-NEXT: %5 = tail call i32 @llvm.x86.sse2.comieq.sd(<2 x double> %2, <2 x double> %4)
+; CHECK-NEXT: ret i32 %5
+ %1 = insertelement <2 x double> undef, double %a, i32 0
+ %2 = insertelement <2 x double> %1, double 1.000000e+00, i32 1
+ %3 = insertelement <2 x double> undef, double %b, i32 0
+ %4 = insertelement <2 x double> %3, double 2.000000e+00, i32 1
+ %5 = tail call i32 @llvm.x86.sse2.comieq.sd(<2 x double> %2, <2 x double> %4)
+ ret i32 %5
+}
+
+define i32 @test_comige_sd_0(double %a, double %b) {
+; CHECK-LABEL: @test_comige_sd_0
+; CHECK-NEXT: %1 = insertelement <2 x double> undef, double %a, i32 0
+; CHECK-NEXT: %2 = insertelement <2 x double> %1, double 1.000000e+00, i32 1
+; CHECK-NEXT: %3 = insertelement <2 x double> undef, double %b, i32 0
+; CHECK-NEXT: %4 = insertelement <2 x double> %3, double 2.000000e+00, i32 1
+; CHECK-NEXT: %5 = tail call i32 @llvm.x86.sse2.comige.sd(<2 x double> %2, <2 x double> %4)
+; CHECK-NEXT: ret i32 %5
+ %1 = insertelement <2 x double> undef, double %a, i32 0
+ %2 = insertelement <2 x double> %1, double 1.000000e+00, i32 1
+ %3 = insertelement <2 x double> undef, double %b, i32 0
+ %4 = insertelement <2 x double> %3, double 2.000000e+00, i32 1
+ %5 = tail call i32 @llvm.x86.sse2.comige.sd(<2 x double> %2, <2 x double> %4)
+ ret i32 %5
+}
+
+define i32 @test_comigt_sd_0(double %a, double %b) {
+; CHECK-LABEL: @test_comigt_sd_0
+; CHECK-NEXT: %1 = insertelement <2 x double> undef, double %a, i32 0
+; CHECK-NEXT: %2 = insertelement <2 x double> %1, double 1.000000e+00, i32 1
+; CHECK-NEXT: %3 = insertelement <2 x double> undef, double %b, i32 0
+; CHECK-NEXT: %4 = insertelement <2 x double> %3, double 2.000000e+00, i32 1
+; CHECK-NEXT: %5 = tail call i32 @llvm.x86.sse2.comigt.sd(<2 x double> %2, <2 x double> %4)
+; CHECK-NEXT: ret i32 %5
+ %1 = insertelement <2 x double> undef, double %a, i32 0
+ %2 = insertelement <2 x double> %1, double 1.000000e+00, i32 1
+ %3 = insertelement <2 x double> undef, double %b, i32 0
+ %4 = insertelement <2 x double> %3, double 2.000000e+00, i32 1
+ %5 = tail call i32 @llvm.x86.sse2.comigt.sd(<2 x double> %2, <2 x double> %4)
+ ret i32 %5
+}
+
+define i32 @test_comile_sd_0(double %a, double %b) {
+; CHECK-LABEL: @test_comile_sd_0
+; CHECK-NEXT: %1 = insertelement <2 x double> undef, double %a, i32 0
+; CHECK-NEXT: %2 = insertelement <2 x double> %1, double 1.000000e+00, i32 1
+; CHECK-NEXT: %3 = insertelement <2 x double> undef, double %b, i32 0
+; CHECK-NEXT: %4 = insertelement <2 x double> %3, double 2.000000e+00, i32 1
+; CHECK-NEXT: %5 = tail call i32 @llvm.x86.sse2.comile.sd(<2 x double> %2, <2 x double> %4)
+; CHECK-NEXT: ret i32 %5
+ %1 = insertelement <2 x double> undef, double %a, i32 0
+ %2 = insertelement <2 x double> %1, double 1.000000e+00, i32 1
+ %3 = insertelement <2 x double> undef, double %b, i32 0
+ %4 = insertelement <2 x double> %3, double 2.000000e+00, i32 1
+ %5 = tail call i32 @llvm.x86.sse2.comile.sd(<2 x double> %2, <2 x double> %4)
+ ret i32 %5
+}
+
+define i32 @test_comilt_sd_0(double %a, double %b) {
+; CHECK-LABEL: @test_comilt_sd_0
+; CHECK-NEXT: %1 = insertelement <2 x double> undef, double %a, i32 0
+; CHECK-NEXT: %2 = insertelement <2 x double> %1, double 1.000000e+00, i32 1
+; CHECK-NEXT: %3 = insertelement <2 x double> undef, double %b, i32 0
+; CHECK-NEXT: %4 = insertelement <2 x double> %3, double 2.000000e+00, i32 1
+; CHECK-NEXT: %5 = tail call i32 @llvm.x86.sse2.comilt.sd(<2 x double> %2, <2 x double> %4)
+; CHECK-NEXT: ret i32 %5
+ %1 = insertelement <2 x double> undef, double %a, i32 0
+ %2 = insertelement <2 x double> %1, double 1.000000e+00, i32 1
+ %3 = insertelement <2 x double> undef, double %b, i32 0
+ %4 = insertelement <2 x double> %3, double 2.000000e+00, i32 1
+ %5 = tail call i32 @llvm.x86.sse2.comilt.sd(<2 x double> %2, <2 x double> %4)
+ ret i32 %5
+}
+
+define i32 @test_comineq_sd_0(double %a, double %b) {
+; CHECK-LABEL: @test_comineq_sd_0
+; CHECK-NEXT: %1 = insertelement <2 x double> undef, double %a, i32 0
+; CHECK-NEXT: %2 = insertelement <2 x double> %1, double 1.000000e+00, i32 1
+; CHECK-NEXT: %3 = insertelement <2 x double> undef, double %b, i32 0
+; CHECK-NEXT: %4 = insertelement <2 x double> %3, double 2.000000e+00, i32 1
+; CHECK-NEXT: %5 = tail call i32 @llvm.x86.sse2.comineq.sd(<2 x double> %2, <2 x double> %4)
+; CHECK-NEXT: ret i32 %5
+ %1 = insertelement <2 x double> undef, double %a, i32 0
+ %2 = insertelement <2 x double> %1, double 1.000000e+00, i32 1
+ %3 = insertelement <2 x double> undef, double %b, i32 0
+ %4 = insertelement <2 x double> %3, double 2.000000e+00, i32 1
+ %5 = tail call i32 @llvm.x86.sse2.comineq.sd(<2 x double> %2, <2 x double> %4)
+ ret i32 %5
+}
+
+define i32 @test_ucomieq_sd_0(double %a, double %b) {
+; CHECK-LABEL: @test_ucomieq_sd_0
+; CHECK-NEXT: %1 = insertelement <2 x double> undef, double %a, i32 0
+; CHECK-NEXT: %2 = insertelement <2 x double> %1, double 1.000000e+00, i32 1
+; CHECK-NEXT: %3 = insertelement <2 x double> undef, double %b, i32 0
+; CHECK-NEXT: %4 = insertelement <2 x double> %3, double 2.000000e+00, i32 1
+; CHECK-NEXT: %5 = tail call i32 @llvm.x86.sse2.ucomieq.sd(<2 x double> %2, <2 x double> %4)
+; CHECK-NEXT: ret i32 %5
+ %1 = insertelement <2 x double> undef, double %a, i32 0
+ %2 = insertelement <2 x double> %1, double 1.000000e+00, i32 1
+ %3 = insertelement <2 x double> undef, double %b, i32 0
+ %4 = insertelement <2 x double> %3, double 2.000000e+00, i32 1
+ %5 = tail call i32 @llvm.x86.sse2.ucomieq.sd(<2 x double> %2, <2 x double> %4)
+ ret i32 %5
+}
+
+define i32 @test_ucomige_sd_0(double %a, double %b) {
+; CHECK-LABEL: @test_ucomige_sd_0
+; CHECK-NEXT: %1 = insertelement <2 x double> undef, double %a, i32 0
+; CHECK-NEXT: %2 = insertelement <2 x double> %1, double 1.000000e+00, i32 1
+; CHECK-NEXT: %3 = insertelement <2 x double> undef, double %b, i32 0
+; CHECK-NEXT: %4 = insertelement <2 x double> %3, double 2.000000e+00, i32 1
+; CHECK-NEXT: %5 = tail call i32 @llvm.x86.sse2.ucomige.sd(<2 x double> %2, <2 x double> %4)
+; CHECK-NEXT: ret i32 %5
+ %1 = insertelement <2 x double> undef, double %a, i32 0
+ %2 = insertelement <2 x double> %1, double 1.000000e+00, i32 1
+ %3 = insertelement <2 x double> undef, double %b, i32 0
+ %4 = insertelement <2 x double> %3, double 2.000000e+00, i32 1
+ %5 = tail call i32 @llvm.x86.sse2.ucomige.sd(<2 x double> %2, <2 x double> %4)
+ ret i32 %5
+}
+
+define i32 @test_ucomigt_sd_0(double %a, double %b) {
+; CHECK-LABEL: @test_ucomigt_sd_0
+; CHECK-NEXT: %1 = insertelement <2 x double> undef, double %a, i32 0
+; CHECK-NEXT: %2 = insertelement <2 x double> %1, double 1.000000e+00, i32 1
+; CHECK-NEXT: %3 = insertelement <2 x double> undef, double %b, i32 0
+; CHECK-NEXT: %4 = insertelement <2 x double> %3, double 2.000000e+00, i32 1
+; CHECK-NEXT: %5 = tail call i32 @llvm.x86.sse2.ucomigt.sd(<2 x double> %2, <2 x double> %4)
+; CHECK-NEXT: ret i32 %5
+ %1 = insertelement <2 x double> undef, double %a, i32 0
+ %2 = insertelement <2 x double> %1, double 1.000000e+00, i32 1
+ %3 = insertelement <2 x double> undef, double %b, i32 0
+ %4 = insertelement <2 x double> %3, double 2.000000e+00, i32 1
+ %5 = tail call i32 @llvm.x86.sse2.ucomigt.sd(<2 x double> %2, <2 x double> %4)
+ ret i32 %5
+}
+
+define i32 @test_ucomile_sd_0(double %a, double %b) {
+; CHECK-LABEL: @test_ucomile_sd_0
+; CHECK-NEXT: %1 = insertelement <2 x double> undef, double %a, i32 0
+; CHECK-NEXT: %2 = insertelement <2 x double> %1, double 1.000000e+00, i32 1
+; CHECK-NEXT: %3 = insertelement <2 x double> undef, double %b, i32 0
+; CHECK-NEXT: %4 = insertelement <2 x double> %3, double 2.000000e+00, i32 1
+; CHECK-NEXT: %5 = tail call i32 @llvm.x86.sse2.ucomile.sd(<2 x double> %2, <2 x double> %4)
+; CHECK-NEXT: ret i32 %5
+ %1 = insertelement <2 x double> undef, double %a, i32 0
+ %2 = insertelement <2 x double> %1, double 1.000000e+00, i32 1
+ %3 = insertelement <2 x double> undef, double %b, i32 0
+ %4 = insertelement <2 x double> %3, double 2.000000e+00, i32 1
+ %5 = tail call i32 @llvm.x86.sse2.ucomile.sd(<2 x double> %2, <2 x double> %4)
+ ret i32 %5
+}
+
+define i32 @test_ucomilt_sd_0(double %a, double %b) {
+; CHECK-LABEL: @test_ucomilt_sd_0
+; CHECK-NEXT: %1 = insertelement <2 x double> undef, double %a, i32 0
+; CHECK-NEXT: %2 = insertelement <2 x double> %1, double 1.000000e+00, i32 1
+; CHECK-NEXT: %3 = insertelement <2 x double> undef, double %b, i32 0
+; CHECK-NEXT: %4 = insertelement <2 x double> %3, double 2.000000e+00, i32 1
+; CHECK-NEXT: %5 = tail call i32 @llvm.x86.sse2.ucomilt.sd(<2 x double> %2, <2 x double> %4)
+; CHECK-NEXT: ret i32 %5
+ %1 = insertelement <2 x double> undef, double %a, i32 0
+ %2 = insertelement <2 x double> %1, double 1.000000e+00, i32 1
+ %3 = insertelement <2 x double> undef, double %b, i32 0
+ %4 = insertelement <2 x double> %3, double 2.000000e+00, i32 1
+ %5 = tail call i32 @llvm.x86.sse2.ucomilt.sd(<2 x double> %2, <2 x double> %4)
+ ret i32 %5
+}
+
+define i32 @test_ucomineq_sd_0(double %a, double %b) {
+; CHECK-LABEL: @test_ucomineq_sd_0
+; CHECK-NEXT: %1 = insertelement <2 x double> undef, double %a, i32 0
+; CHECK-NEXT: %2 = insertelement <2 x double> %1, double 1.000000e+00, i32 1
+; CHECK-NEXT: %3 = insertelement <2 x double> undef, double %b, i32 0
+; CHECK-NEXT: %4 = insertelement <2 x double> %3, double 2.000000e+00, i32 1
+; CHECK-NEXT: %5 = tail call i32 @llvm.x86.sse2.ucomineq.sd(<2 x double> %2, <2 x double> %4)
+; CHECK-NEXT: ret i32 %5
+ %1 = insertelement <2 x double> undef, double %a, i32 0
+ %2 = insertelement <2 x double> %1, double 1.000000e+00, i32 1
+ %3 = insertelement <2 x double> undef, double %b, i32 0
+ %4 = insertelement <2 x double> %3, double 2.000000e+00, i32 1
+ %5 = tail call i32 @llvm.x86.sse2.ucomineq.sd(<2 x double> %2, <2 x double> %4)
+ ret i32 %5
+}
+
declare <2 x double> @llvm.x86.sse2.sqrt.sd(<2 x double>) nounwind readnone
declare <2 x double> @llvm.x86.sse2.add.sd(<2 x double>, <2 x double>)
@@ -136,3 +328,17 @@ declare <2 x double> @llvm.x86.sse2.div.sd(<2 x double>, <2 x double>)
declare <2 x double> @llvm.x86.sse2.min.sd(<2 x double>, <2 x double>)
declare <2 x double> @llvm.x86.sse2.max.sd(<2 x double>, <2 x double>)
declare <2 x double> @llvm.x86.sse2.cmp.sd(<2 x double>, <2 x double>, i8)
+
+declare i32 @llvm.x86.sse2.comieq.sd(<2 x double>, <2 x double>)
+declare i32 @llvm.x86.sse2.comige.sd(<2 x double>, <2 x double>)
+declare i32 @llvm.x86.sse2.comigt.sd(<2 x double>, <2 x double>)
+declare i32 @llvm.x86.sse2.comile.sd(<2 x double>, <2 x double>)
+declare i32 @llvm.x86.sse2.comilt.sd(<2 x double>, <2 x double>)
+declare i32 @llvm.x86.sse2.comineq.sd(<2 x double>, <2 x double>)
+
+declare i32 @llvm.x86.sse2.ucomieq.sd(<2 x double>, <2 x double>)
+declare i32 @llvm.x86.sse2.ucomige.sd(<2 x double>, <2 x double>)
+declare i32 @llvm.x86.sse2.ucomigt.sd(<2 x double>, <2 x double>)
+declare i32 @llvm.x86.sse2.ucomile.sd(<2 x double>, <2 x double>)
+declare i32 @llvm.x86.sse2.ucomilt.sd(<2 x double>, <2 x double>)
+declare i32 @llvm.x86.sse2.ucomineq.sd(<2 x double>, <2 x double>)
OpenPOWER on IntegriCloud