[llvm] 7e1f376 - [InstCombine] add tests for integer reductions; NFC
Sanjay Patel via llvm-commits
llvm-commits at lists.llvm.org
Mon Jun 22 18:05:10 PDT 2020
Author: Sanjay Patel
Date: 2020-06-22T20:47:09-04:00
New Revision: 7e1f376f8016bcfea4067de28c42ec6a7eb091fe
URL: https://github.com/llvm/llvm-project/commit/7e1f376f8016bcfea4067de28c42ec6a7eb091fe
DIFF: https://github.com/llvm/llvm-project/commit/7e1f376f8016bcfea4067de28c42ec6a7eb091fe.diff
LOG: [InstCombine] add tests for integer reductions; NFC
Added:
Modified:
llvm/test/Transforms/InstCombine/vector-reductions.ll
Removed:
################################################################################
diff --git a/llvm/test/Transforms/InstCombine/vector-reductions.ll b/llvm/test/Transforms/InstCombine/vector-reductions.ll
index dfac8da97b35..bd051eb830a3 100644
--- a/llvm/test/Transforms/InstCombine/vector-reductions.ll
+++ b/llvm/test/Transforms/InstCombine/vector-reductions.ll
@@ -5,6 +5,10 @@ declare float @llvm.experimental.vector.reduce.v2.fadd.f32.v4f32(float, <4 x flo
declare float @llvm.experimental.vector.reduce.v2.fadd.f32.v8f32(float, <8 x float>)
declare void @use_f32(float)
+declare i32 @llvm.experimental.vector.reduce.add.v4i32(<4 x i32>)
+declare i32 @llvm.experimental.vector.reduce.add.v8i32(<8 x i32>)
+declare void @use_i32(i32)
+
define float @
diff _of_sums_v4f32(float %a0, <4 x float> %v0, float %a1, <4 x float> %v1) {
; CHECK-LABEL: @
diff _of_sums_v4f32(
; CHECK-NEXT: [[TMP1:%.*]] = fsub reassoc nsz <4 x float> [[V0:%.*]], [[V1:%.*]]
@@ -81,3 +85,59 @@ define float @
diff _of_sums_type_mismatch(float %a0, <4 x float> %v0, float %a1,
%r = fsub fast float %r0, %r1
ret float %r
}
+
+define i32 @
diff _of_sums_v4i32(<4 x i32> %v0, <4 x i32> %v1) {
+; CHECK-LABEL: @
diff _of_sums_v4i32(
+; CHECK-NEXT: [[R0:%.*]] = call i32 @llvm.experimental.vector.reduce.add.v4i32(<4 x i32> [[V0:%.*]])
+; CHECK-NEXT: [[R1:%.*]] = call i32 @llvm.experimental.vector.reduce.add.v4i32(<4 x i32> [[V1:%.*]])
+; CHECK-NEXT: [[R:%.*]] = sub i32 [[R0]], [[R1]]
+; CHECK-NEXT: ret i32 [[R]]
+;
+ %r0 = call i32 @llvm.experimental.vector.reduce.add.v4i32(<4 x i32> %v0)
+ %r1 = call i32 @llvm.experimental.vector.reduce.add.v4i32(<4 x i32> %v1)
+ %r = sub i32 %r0, %r1
+ ret i32 %r
+}
+
+define i32 @
diff _of_sums_v4i32_extra_use1(<4 x i32> %v0, <4 x i32> %v1) {
+; CHECK-LABEL: @
diff _of_sums_v4i32_extra_use1(
+; CHECK-NEXT: [[R0:%.*]] = call i32 @llvm.experimental.vector.reduce.add.v4i32(<4 x i32> [[V0:%.*]])
+; CHECK-NEXT: call void @use_i32(i32 [[R0]])
+; CHECK-NEXT: [[R1:%.*]] = call i32 @llvm.experimental.vector.reduce.add.v4i32(<4 x i32> [[V1:%.*]])
+; CHECK-NEXT: [[R:%.*]] = sub i32 [[R0]], [[R1]]
+; CHECK-NEXT: ret i32 [[R]]
+;
+ %r0 = call i32 @llvm.experimental.vector.reduce.add.v4i32(<4 x i32> %v0)
+ call void @use_i32(i32 %r0)
+ %r1 = call i32 @llvm.experimental.vector.reduce.add.v4i32(<4 x i32> %v1)
+ %r = sub i32 %r0, %r1
+ ret i32 %r
+}
+
+define i32 @
diff _of_sums_v4i32_extra_use2(<4 x i32> %v0, <4 x i32> %v1) {
+; CHECK-LABEL: @
diff _of_sums_v4i32_extra_use2(
+; CHECK-NEXT: [[R0:%.*]] = call i32 @llvm.experimental.vector.reduce.add.v4i32(<4 x i32> [[V0:%.*]])
+; CHECK-NEXT: [[R1:%.*]] = call i32 @llvm.experimental.vector.reduce.add.v4i32(<4 x i32> [[V1:%.*]])
+; CHECK-NEXT: call void @use_i32(i32 [[R1]])
+; CHECK-NEXT: [[R:%.*]] = sub i32 [[R0]], [[R1]]
+; CHECK-NEXT: ret i32 [[R]]
+;
+ %r0 = call i32 @llvm.experimental.vector.reduce.add.v4i32(<4 x i32> %v0)
+ %r1 = call i32 @llvm.experimental.vector.reduce.add.v4i32(<4 x i32> %v1)
+ call void @use_i32(i32 %r1)
+ %r = sub i32 %r0, %r1
+ ret i32 %r
+}
+
+define i32 @
diff _of_sums_type_mismatch2(<8 x i32> %v0, <4 x i32> %v1) {
+; CHECK-LABEL: @
diff _of_sums_type_mismatch2(
+; CHECK-NEXT: [[R0:%.*]] = call i32 @llvm.experimental.vector.reduce.add.v8i32(<8 x i32> [[V0:%.*]])
+; CHECK-NEXT: [[R1:%.*]] = call i32 @llvm.experimental.vector.reduce.add.v4i32(<4 x i32> [[V1:%.*]])
+; CHECK-NEXT: [[R:%.*]] = sub i32 [[R0]], [[R1]]
+; CHECK-NEXT: ret i32 [[R]]
+;
+ %r0 = call i32 @llvm.experimental.vector.reduce.add.v8i32(<8 x i32> %v0)
+ %r1 = call i32 @llvm.experimental.vector.reduce.add.v4i32(<4 x i32> %v1)
+ %r = sub i32 %r0, %r1
+ ret i32 %r
+}
More information about the llvm-commits
mailing list