[SLPVectorize] Basic ephemeral-value awareness

[oota-llvm.git] / test / Transforms / SLPVectorizer / X86 / insert-element-build-vector.ll

diff --git a/test/Transforms/SLPVectorizer/X86/insert-element-build-vector.ll b/test/Transforms/SLPVectorizer/X86/insert-element-build-vector.ll
index 620292bdaf385e034b8755f382f39222f59781e6..0221613b9707f9b3edb2e2e60fea4820f08208c5 100644 (file)
--- a/test/Transforms/SLPVectorizer/X86/insert-element-build-vector.ll
+++ b/test/Transforms/SLPVectorizer/X86/insert-element-build-vector.ll
@@ -35,6 +35,49 @@ define <4 x float> @simple_select(<4 x float> %a, <4 x float> %b, <4 x i32> %c)
   ret <4 x float> %rd
 }
 
+declare void @llvm.assume(i1) nounwind
+
+; This entire tree is ephemeral, don't vectorize any of it.
+define <4 x float> @simple_select_eph(<4 x float> %a, <4 x float> %b, <4 x i32> %c) #0 {
+; CHECK-LABEL: @simple_select_eph(
+; CHECK-NOT: icmp ne <4 x i32>
+; CHECK-NOT: select <4 x i1>
+  %c0 = extractelement <4 x i32> %c, i32 0
+  %c1 = extractelement <4 x i32> %c, i32 1
+  %c2 = extractelement <4 x i32> %c, i32 2
+  %c3 = extractelement <4 x i32> %c, i32 3
+  %a0 = extractelement <4 x float> %a, i32 0
+  %a1 = extractelement <4 x float> %a, i32 1
+  %a2 = extractelement <4 x float> %a, i32 2
+  %a3 = extractelement <4 x float> %a, i32 3
+  %b0 = extractelement <4 x float> %b, i32 0
+  %b1 = extractelement <4 x float> %b, i32 1
+  %b2 = extractelement <4 x float> %b, i32 2
+  %b3 = extractelement <4 x float> %b, i32 3
+  %cmp0 = icmp ne i32 %c0, 0
+  %cmp1 = icmp ne i32 %c1, 0
+  %cmp2 = icmp ne i32 %c2, 0
+  %cmp3 = icmp ne i32 %c3, 0
+  %s0 = select i1 %cmp0, float %a0, float %b0
+  %s1 = select i1 %cmp1, float %a1, float %b1
+  %s2 = select i1 %cmp2, float %a2, float %b2
+  %s3 = select i1 %cmp3, float %a3, float %b3
+  %ra = insertelement <4 x float> undef, float %s0, i32 0
+  %rb = insertelement <4 x float> %ra, float %s1, i32 1
+  %rc = insertelement <4 x float> %rb, float %s2, i32 2
+  %rd = insertelement <4 x float> %rc, float %s3, i32 3
+  %q0 = extractelement <4 x float> %rd, i32 0
+  %q1 = extractelement <4 x float> %rd, i32 1
+  %q2 = extractelement <4 x float> %rd, i32 2
+  %q3 = extractelement <4 x float> %rd, i32 3
+  %q4 = fadd float %q0, %q1
+  %q5 = fadd float %q2, %q3
+  %q6 = fadd float %q4, %q5
+  %qi = fcmp olt float %q6, %q5
+  call void @llvm.assume(i1 %qi)
+  ret <4 x float> undef
+}
+
 ; Insert in an order different from the vector indices to make sure it
 ; doesn't matter
 define <4 x float> @simple_select_insert_out_of_order(<4 x float> %a, <4 x float> %b, <4 x i32> %c) #0 {
@@ -223,7 +266,7 @@ define <4 x float> @reschedule_extract(<4 x float> %a, <4 x float> %b) {
 ; instructions that are erased.
 define <4 x float> @take_credit(<4 x float> %a, <4 x float> %b) {
 ; ZEROTHRESH-LABEL: @take_credit(
-; ZEROTHRESH-CHECK: %1 = fadd <4 x float> %a, %b
+; ZEROTHRESH: %1 = fadd <4 x float> %a, %b
   %a0 = extractelement <4 x float> %a, i32 0
   %b0 = extractelement <4 x float> %b, i32 0
   %c0 = fadd float %a0, %b0
@@ -243,4 +286,40 @@ define <4 x float> @take_credit(<4 x float> %a, <4 x float> %b) {
   ret <4 x float> %v3
 }
 
+; Make sure we handle multiple trees that feed one build vector correctly.
+define <4 x double> @multi_tree(double %w, double %x, double %y, double %z) {
+entry:
+  %t0 = fadd double %w , 0.000000e+00
+  %t1 = fadd double %x , 1.000000e+00
+  %t2 = fadd double %y , 2.000000e+00
+  %t3 = fadd double %z , 3.000000e+00
+  %t4 = fmul double %t0, 1.000000e+00
+  %i1 = insertelement <4 x double> undef, double %t4, i32 3
+  %t5 = fmul double %t1, 1.000000e+00
+  %i2 = insertelement <4 x double> %i1, double %t5, i32 2
+  %t6 = fmul double %t2, 1.000000e+00
+  %i3 = insertelement <4 x double> %i2, double %t6, i32 1
+  %t7 = fmul double %t3, 1.000000e+00
+  %i4 = insertelement <4 x double> %i3, double %t7, i32 0
+  ret <4 x double> %i4
+}
+; CHECK-LABEL: @multi_tree
+; CHECK-DAG:  %[[V0:.+]] = insertelement <2 x double> undef, double %w, i32 0
+; CHECK-DAG:  %[[V1:.+]] = insertelement <2 x double> %[[V0]], double %x, i32 1
+; CHECK-DAG:  %[[V2:.+]] = fadd <2 x double> %[[V1]], <double 0.000000e+00, double 1.000000e+00>
+; CHECK-DAG:  %[[V3:.+]] = insertelement <2 x double> undef, double %y, i32 0
+; CHECK-DAG:  %[[V4:.+]] = insertelement <2 x double> %[[V3]], double %z, i32 1
+; CHECK-DAG:  %[[V5:.+]] = fadd <2 x double> %[[V4]], <double 2.000000e+00, double 3.000000e+00>
+; CHECK-DAG:  %[[V6:.+]] = fmul <2 x double> <double 1.000000e+00, double 1.000000e+00>, %[[V2]]
+; CHECK-DAG:  %[[V7:.+]] = extractelement <2 x double> %[[V6]], i32 0
+; CHECK-DAG:  %[[I1:.+]] = insertelement <4 x double> undef, double %[[V7]], i32 3
+; CHECK-DAG:  %[[V8:.+]] = extractelement <2 x double> %[[V6]], i32 1
+; CHECK-DAG:  %[[I2:.+]] = insertelement <4 x double> %[[I1]], double %[[V8]], i32 2
+; CHECK-DAG:  %[[V9:.+]] = fmul <2 x double> <double 1.000000e+00, double 1.000000e+00>, %[[V5]]
+; CHECK-DAG:  %[[V10:.+]] = extractelement <2 x double> %[[V9]], i32 0
+; CHECK-DAG:  %[[I3:.+]] = insertelement <4 x double> %i2, double %[[V10]], i32 1
+; CHECK-DAG:  %[[V11:.+]] = extractelement <2 x double> %[[V9]], i32 1
+; CHECK-DAG:  %[[I4:.+]] = insertelement <4 x double> %i3, double %[[V11]], i32 0
+; CHECK:  ret <4 x double> %[[I4]]
+
 attributes #0 = { nounwind ssp uwtable "less-precise-fpmad"="false" "no-frame-pointer-elim"="true" "no-frame-pointer-elim-non-leaf"="true" "no-infs-fp-math"="false" "no-nans-fp-math"="false" "stack-protector-buffer-size"="8" "unsafe-fp-math"="false" "use-soft-float"="false" }