[CIR][NFC] Move some builtin tests to the CodeGenBuitins folder - #183607
Merged
Merged
Conversation
ayokunle321
changed the title
llvmbot
added
clang
Member
|
@llvm/pr-subscribers-clangir @llvm/pr-subscribers-clang Author: Ayokunle Amodu (ayokunle321) ChangesThis moves a few tests that were created in the wrong location. Patch is 52.04 KiB, truncated to 20.00 KiB below, full version: https://github.com/llvm/llvm-project/pull/183607.diff 5 Files Affected:
diff --git a/clang/test/CIR/CodeGen/builtins-elementwise.c b/clang/test/CIR/CodeGen/builtins-elementwise.c
deleted file mode 100644
index 6c31ce19788d9..0000000000000
--- a/clang/test/CIR/CodeGen/builtins-elementwise.c
+++ /dev/null
@@ -1,515 +0,0 @@
-// RUN: %clang_cc1 -triple aarch64-none-linux-android24 -emit-cir %s -o %t.cir
-// RUN: FileCheck --check-prefix=CIR --input-file=%t.cir %s
-// RUN: %clang_cc1 -triple aarch64-none-linux-android24 -fclangir \
-// RUN: -emit-llvm %s -o %t.ll
-// RUN: FileCheck --check-prefix=LLVM --input-file=%t.ll %s
-// RUN: %clang_cc1 -triple aarch64-none-linux-android24 -emit-llvm %s -o %t-ogcg.ll
-// RUN: FileCheck --check-prefix=OGCG --input-file=%t-ogcg.ll %s
-
-typedef int vint4 __attribute__((ext_vector_type(4)));
-typedef float vfloat4 __attribute__((ext_vector_type(4)));
-typedef double vdouble4 __attribute__((ext_vector_type(4)));
-
-void test_builtin_elementwise_abs(vint4 vi4, int i, float f, double d,
- vfloat4 vf4, vdouble4 vd4) {
- // CIR-LABEL: test_builtin_elementwise_abs
- // LLVM-LABEL: test_builtin_elementwise_abs
- // OGCG-LABEL: test_builtin_elementwise_abs
- // CIR: {{%.*}} = cir.fabs {{%.*}} : !cir.float
- // LLVM: {{%.*}} = call float @llvm.fabs.f32(float {{%.*}})
- // OGCG: {{%.*}} = call float @llvm.fabs.f32(float {{%.*}})
- f = __builtin_elementwise_abs(f);
-
- // CIR: {{%.*}} = cir.fabs {{%.*}} : !cir.double
- // LLVM: {{%.*}} = call double @llvm.fabs.f64(double {{%.*}})
- // OGCG: {{%.*}} = call double @llvm.fabs.f64(double {{%.*}})
- d = __builtin_elementwise_abs(d);
-
- // CIR: {{%.*}} = cir.abs {{%.*}} : !cir.vector<4 x !s32i>
- // LLVM: {{%.*}} = call <4 x i32> @llvm.abs.v4i32(<4 x i32> {{%.*}}, i1 false)
- // OGCG: {{%.*}} = call <4 x i32> @llvm.abs.v4i32(<4 x i32> {{%.*}}, i1 false)
- vi4 = __builtin_elementwise_abs(vi4);
-
- // CIR: {{%.*}} = cir.abs {{%.*}} : !s32
- // LLVM: {{%.*}} = call i32 @llvm.abs.i32(i32 {{%.*}}, i1 false)
- // OGCG: {{%.*}} = call i32 @llvm.abs.i32(i32 {{%.*}}, i1 false)
- i = __builtin_elementwise_abs(i);
-
- // CIR: {{%.*}} = cir.fabs {{%.*}} : !cir.vector<4 x !cir.float>
- // LLVM: {{%.*}} = call <4 x float> @llvm.fabs.v4f32(<4 x float> {{%.*}})
- // OGCG: {{%.*}} = call <4 x float> @llvm.fabs.v4f32(<4 x float> {{%.*}})
- vf4 = __builtin_elementwise_abs(vf4);
-
- // CIR: {{%.*}} = cir.fabs {{%.*}} : !cir.vector<4 x !cir.double>
- // LLVM: {{%.*}} = call <4 x double> @llvm.fabs.v4f64(<4 x double> {{%.*}})
- // OGCG: {{%.*}} = call <4 x double> @llvm.fabs.v4f64(<4 x double> {{%.*}})
- vd4 = __builtin_elementwise_abs(vd4);
-}
-
-void test_builtin_elementwise_acos(float f, double d, vfloat4 vf4,
- vdouble4 vd4) {
- // CIR-LABEL: test_builtin_elementwise_acos
- // LLVM-LABEL: test_builtin_elementwise_acos
- // OGCG-LABEL: test_builtin_elementwise_acos
- // CIR: {{%.*}} = cir.acos {{%.*}} : !cir.float
- // LLVM: {{%.*}} = call float @llvm.acos.f32(float {{%.*}})
- // OGCG: {{%.*}} = call float @llvm.acos.f32(float {{%.*}})
- f = __builtin_elementwise_acos(f);
-
- // CIR: {{%.*}} = cir.acos {{%.*}} : !cir.double
- // LLVM: {{%.*}} = call double @llvm.acos.f64(double {{%.*}})
- // OGCG: {{%.*}} = call double @llvm.acos.f64(double {{%.*}})
- d = __builtin_elementwise_acos(d);
-
- // CIR: {{%.*}} = cir.acos {{%.*}} : !cir.vector<4 x !cir.float>
- // LLVM: {{%.*}} = call <4 x float> @llvm.acos.v4f32(<4 x float> {{%.*}})
- // OGCG: {{%.*}} = call <4 x float> @llvm.acos.v4f32(<4 x float> {{%.*}})
- vf4 = __builtin_elementwise_acos(vf4);
-
- // CIR: {{%.*}} = cir.acos {{%.*}} : !cir.vector<4 x !cir.double>
- // LLVM: {{%.*}} = call <4 x double> @llvm.acos.v4f64(<4 x double> {{%.*}})
- // OGCG: {{%.*}} = call <4 x double> @llvm.acos.v4f64(<4 x double> {{%.*}})
- vd4 = __builtin_elementwise_acos(vd4);
-}
-
-void test_builtin_elementwise_asin(float f, double d, vfloat4 vf4,
- vdouble4 vd4) {
- // CIR-LABEL: test_builtin_elementwise_asin
- // LLVM-LABEL: test_builtin_elementwise_asin
- // OGCG-LABEL: test_builtin_elementwise_asin
- // CIR: {{%.*}} = cir.asin {{%.*}} : !cir.float
- // LLVM: {{%.*}} = call float @llvm.asin.f32(float {{%.*}})
- // OGCG: {{%.*}} = call float @llvm.asin.f32(float {{%.*}})
- f = __builtin_elementwise_asin(f);
-
- // CIR: {{%.*}} = cir.asin {{%.*}} : !cir.double
- // LLVM: {{%.*}} = call double @llvm.asin.f64(double {{%.*}})
- // OGCG: {{%.*}} = call double @llvm.asin.f64(double {{%.*}})
- d = __builtin_elementwise_asin(d);
-
- // CIR: {{%.*}} = cir.asin {{%.*}} : !cir.vector<4 x !cir.float>
- // LLVM: {{%.*}} = call <4 x float> @llvm.asin.v4f32(<4 x float> {{%.*}})
- // OGCG: {{%.*}} = call <4 x float> @llvm.asin.v4f32(<4 x float> {{%.*}})
- vf4 = __builtin_elementwise_asin(vf4);
-
- // CIR: {{%.*}} = cir.asin {{%.*}} : !cir.vector<4 x !cir.double>
- // LLVM: {{%.*}} = call <4 x double> @llvm.asin.v4f64(<4 x double> {{%.*}})
- // OGCG: {{%.*}} = call <4 x double> @llvm.asin.v4f64(<4 x double> {{%.*}})
- vd4 = __builtin_elementwise_asin(vd4);
-}
-
-void test_builtin_elementwise_atan(float f, double d, vfloat4 vf4,
- vdouble4 vd4) {
- // CIR-LABEL: test_builtin_elementwise_atan
- // LLVM-LABEL: test_builtin_elementwise_atan
- // OGCG-LABEL: test_builtin_elementwise_atan
- // CIR: {{%.*}} = cir.atan {{%.*}} : !cir.float
- // LLVM: {{%.*}} = call float @llvm.atan.f32(float {{%.*}})
- // OGCG: {{%.*}} = call float @llvm.atan.f32(float {{%.*}})
- f = __builtin_elementwise_atan(f);
-
- // CIR: {{%.*}} = cir.atan {{%.*}} : !cir.double
- // LLVM: {{%.*}} = call double @llvm.atan.f64(double {{%.*}})
- // OGCG: {{%.*}} = call double @llvm.atan.f64(double {{%.*}})
- d = __builtin_elementwise_atan(d);
-
- // CIR: {{%.*}} = cir.atan {{%.*}} : !cir.vector<4 x !cir.float>
- // LLVM: {{%.*}} = call <4 x float> @llvm.atan.v4f32(<4 x float> {{%.*}})
- // OGCG: {{%.*}} = call <4 x float> @llvm.atan.v4f32(<4 x float> {{%.*}})
- vf4 = __builtin_elementwise_atan(vf4);
-
- // CIR: {{%.*}} = cir.atan {{%.*}} : !cir.vector<4 x !cir.double>
- // LLVM: {{%.*}} = call <4 x double> @llvm.atan.v4f64(<4 x double> {{%.*}})
- // OGCG: {{%.*}} = call <4 x double> @llvm.atan.v4f64(<4 x double> {{%.*}})
- vd4 = __builtin_elementwise_atan(vd4);
-}
-
-void test_builtin_elementwise_atan2(float f, double d, vfloat4 vf4,
- vdouble4 vd4) {
- // CIR-LABEL: test_builtin_elementwise_atan2
- // LLVM-LABEL: test_builtin_elementwise_atan2
- // OGCG-LABEL: test_builtin_elementwise_atan2
- // CIR: {{%.*}} = cir.atan2 {{%.*}}, {{%.*}} : !cir.float
- // LLVM: {{%.*}} = call float @llvm.atan2.f32(float {{%.*}}, float {{%.*}})
- // OGCG: {{%.*}} = call float @llvm.atan2.f32(float {{%.*}}, float {{%.*}})
- f = __builtin_elementwise_atan2(f, f);
-
- // CIR: {{%.*}} = cir.atan2 {{%.*}}, {{%.*}} : !cir.double
- // LLVM: {{%.*}} = call double @llvm.atan2.f64(double {{%.*}}, double {{%.*}})
- // OGCG: {{%.*}} = call double @llvm.atan2.f64(double {{%.*}}, double {{%.*}})
- d = __builtin_elementwise_atan2(d, d);
-
- // CIR: {{%.*}} = cir.atan2 {{%.*}}, {{%.*}} : !cir.vector<4 x !cir.float>
- // LLVM: {{%.*}} = call <4 x float> @llvm.atan2.v4f32(<4 x float> {{%.*}}, <4 x float> {{%.*}})
- // OGCG: {{%.*}} = call <4 x float> @llvm.atan2.v4f32(<4 x float> {{%.*}}, <4 x float> {{%.*}})
- vf4 = __builtin_elementwise_atan2(vf4, vf4);
-
- // CIR: {{%.*}} = cir.atan2 {{%.*}}, {{%.*}} : !cir.vector<4 x !cir.double>
- // LLVM: {{%.*}} = call <4 x double> @llvm.atan2.v4f64(<4 x double> {{%.*}}, <4 x double> {{%.*}})
- // OGCG: {{%.*}} = call <4 x double> @llvm.atan2.v4f64(<4 x double> {{%.*}}, <4 x double> {{%.*}})
- vd4 = __builtin_elementwise_atan2(vd4, vd4);
-}
-
-void test_builtin_elementwise_exp(float f, double d, vfloat4 vf4,
- vdouble4 vd4) {
- // CIR-LABEL: test_builtin_elementwise_exp
- // LLVM-LABEL: test_builtin_elementwise_exp
- // OGCG-LABEL: test_builtin_elementwise_exp
- // CIR: {{%.*}} = cir.exp {{%.*}} : !cir.float
- // LLVM: {{%.*}} = call float @llvm.exp.f32(float {{%.*}})
- // OGCG: {{%.*}} = call float @llvm.exp.f32(float {{%.*}})
- f = __builtin_elementwise_exp(f);
-
- // CIR: {{%.*}} = cir.exp {{%.*}} : !cir.double
- // LLVM: {{%.*}} = call double @llvm.exp.f64(double {{%.*}})
- // OGCG: {{%.*}} = call double @llvm.exp.f64(double {{%.*}})
- d = __builtin_elementwise_exp(d);
-
- // CIR: {{%.*}} = cir.exp {{%.*}} : !cir.vector<4 x !cir.float>
- // LLVM: {{%.*}} = call <4 x float> @llvm.exp.v4f32(<4 x float> {{%.*}})
- // OGCG: {{%.*}} = call <4 x float> @llvm.exp.v4f32(<4 x float> {{%.*}})
- vf4 = __builtin_elementwise_exp(vf4);
-
- // CIR: {{%.*}} = cir.exp {{%.*}} : !cir.vector<4 x !cir.double>
- // LLVM: {{%.*}} = call <4 x double> @llvm.exp.v4f64(<4 x double> {{%.*}})
- // OGCG: {{%.*}} = call <4 x double> @llvm.exp.v4f64(<4 x double> {{%.*}})
- vd4 = __builtin_elementwise_exp(vd4);
-}
-
-void test_builtin_elementwise_exp2(float f, double d, vfloat4 vf4,
- vdouble4 vd4) {
- // CIR-LABEL: test_builtin_elementwise_exp2
- // LLVM-LABEL: test_builtin_elementwise_exp2
- // OGCG-LABEL: test_builtin_elementwise_exp2
- // CIR: {{%.*}} = cir.exp2 {{%.*}} : !cir.float
- // LLVM: {{%.*}} = call float @llvm.exp2.f32(float {{%.*}})
- // OGCG: {{%.*}} = call float @llvm.exp2.f32(float {{%.*}})
- f = __builtin_elementwise_exp2(f);
-
- // CIR: {{%.*}} = cir.exp2 {{%.*}} : !cir.double
- // LLVM: {{%.*}} = call double @llvm.exp2.f64(double {{%.*}})
- // OGCG: {{%.*}} = call double @llvm.exp2.f64(double {{%.*}})
- d = __builtin_elementwise_exp2(d);
-
- // CIR: {{%.*}} = cir.exp2 {{%.*}} : !cir.vector<4 x !cir.float>
- // LLVM: {{%.*}} = call <4 x float> @llvm.exp2.v4f32(<4 x float> {{%.*}})
- // OGCG: {{%.*}} = call <4 x float> @llvm.exp2.v4f32(<4 x float> {{%.*}})
- vf4 = __builtin_elementwise_exp2(vf4);
-
- // CIR: {{%.*}} = cir.exp2 {{%.*}} : !cir.vector<4 x !cir.double>
- // LLVM: {{%.*}} = call <4 x double> @llvm.exp2.v4f64(<4 x double> {{%.*}})
- // OGCG: {{%.*}} = call <4 x double> @llvm.exp2.v4f64(<4 x double> {{%.*}})
- vd4 = __builtin_elementwise_exp2(vd4);
-}
-
-void test_builtin_elementwise_log(float f, double d, vfloat4 vf4,
- vdouble4 vd4) {
- // CIR-LABEL: test_builtin_elementwise_log
- // LLVM-LABEL: test_builtin_elementwise_log
- // OGCG-LABEL: test_builtin_elementwise_log
- // CIR: {{%.*}} = cir.log {{%.*}} : !cir.float
- // LLVM: {{%.*}} = call float @llvm.log.f32(float {{%.*}})
- // OGCG: {{%.*}} = call float @llvm.log.f32(float {{%.*}})
- f = __builtin_elementwise_log(f);
-
- // CIR: {{%.*}} = cir.log {{%.*}} : !cir.double
- // LLVM: {{%.*}} = call double @llvm.log.f64(double {{%.*}})
- // OGCG: {{%.*}} = call double @llvm.log.f64(double {{%.*}})
- d = __builtin_elementwise_log(d);
-
- // CIR: {{%.*}} = cir.log {{%.*}} : !cir.vector<4 x !cir.float>
- // LLVM: {{%.*}} = call <4 x float> @llvm.log.v4f32(<4 x float> {{%.*}})
- // OGCG: {{%.*}} = call <4 x float> @llvm.log.v4f32(<4 x float> {{%.*}})
- vf4 = __builtin_elementwise_log(vf4);
-
- // CIR: {{%.*}} = cir.log {{%.*}} : !cir.vector<4 x !cir.double>
- // LLVM: {{%.*}} = call <4 x double> @llvm.log.v4f64(<4 x double> {{%.*}})
- // OGCG: {{%.*}} = call <4 x double> @llvm.log.v4f64(<4 x double> {{%.*}})
- vd4 = __builtin_elementwise_log(vd4);
-}
-
-void test_builtin_elementwise_log2(float f, double d, vfloat4 vf4,
- vdouble4 vd4) {
- // CIR-LABEL: test_builtin_elementwise_log2
- // LLVM-LABEL: test_builtin_elementwise_log2
- // OGCG-LABEL: test_builtin_elementwise_log2
- // CIR: {{%.*}} = cir.log2 {{%.*}} : !cir.float
- // LLVM: {{%.*}} = call float @llvm.log2.f32(float {{%.*}})
- // OGCG: {{%.*}} = call float @llvm.log2.f32(float {{%.*}})
- f = __builtin_elementwise_log2(f);
-
- // CIR: {{%.*}} = cir.log2 {{%.*}} : !cir.double
- // LLVM: {{%.*}} = call double @llvm.log2.f64(double {{%.*}})
- // OGCG: {{%.*}} = call double @llvm.log2.f64(double {{%.*}})
- d = __builtin_elementwise_log2(d);
-
- // CIR: {{%.*}} = cir.log2 {{%.*}} : !cir.vector<4 x !cir.float>
- // LLVM: {{%.*}} = call <4 x float> @llvm.log2.v4f32(<4 x float> {{%.*}})
- // OGCG: {{%.*}} = call <4 x float> @llvm.log2.v4f32(<4 x float> {{%.*}})
- vf4 = __builtin_elementwise_log2(vf4);
-
- // CIR: {{%.*}} = cir.log2 {{%.*}} : !cir.vector<4 x !cir.double>
- // LLVM: {{%.*}} = call <4 x double> @llvm.log2.v4f64(<4 x double> {{%.*}})
- // OGCG: {{%.*}} = call <4 x double> @llvm.log2.v4f64(<4 x double> {{%.*}})
- vd4 = __builtin_elementwise_log2(vd4);
-}
-
-void test_builtin_elementwise_log10(float f, double d, vfloat4 vf4,
- vdouble4 vd4) {
- // CIR-LABEL: test_builtin_elementwise_log10
- // LLVM-LABEL: test_builtin_elementwise_log10
- // OGCG-LABEL: test_builtin_elementwise_log10
- // CIR: {{%.*}} = cir.log10 {{%.*}} : !cir.float
- // LLVM: {{%.*}} = call float @llvm.log10.f32(float {{%.*}})
- // OGCG: {{%.*}} = call float @llvm.log10.f32(float {{%.*}})
- f = __builtin_elementwise_log10(f);
-
- // CIR: {{%.*}} = cir.log10 {{%.*}} : !cir.double
- // LLVM: {{%.*}} = call double @llvm.log10.f64(double {{%.*}})
- // OGCG: {{%.*}} = call double @llvm.log10.f64(double {{%.*}})
- d = __builtin_elementwise_log10(d);
-
- // CIR: {{%.*}} = cir.log10 {{%.*}} : !cir.vector<4 x !cir.float>
- // LLVM: {{%.*}} = call <4 x float> @llvm.log10.v4f32(<4 x float> {{%.*}})
- // OGCG: {{%.*}} = call <4 x float> @llvm.log10.v4f32(<4 x float> {{%.*}})
- vf4 = __builtin_elementwise_log10(vf4);
-
- // CIR: {{%.*}} = cir.log10 {{%.*}} : !cir.vector<4 x !cir.double>
- // LLVM: {{%.*}} = call <4 x double> @llvm.log10.v4f64(<4 x double> {{%.*}})
- // OGCG: {{%.*}} = call <4 x double> @llvm.log10.v4f64(<4 x double> {{%.*}})
- vd4 = __builtin_elementwise_log10(vd4);
-}
-
-void test_builtin_elementwise_cos(float f, double d, vfloat4 vf4,
- vdouble4 vd4) {
- // CIR-LABEL: test_builtin_elementwise_cos
- // LLVM-LABEL: test_builtin_elementwise_cos
- // OGCG-LABEL: test_builtin_elementwise_cos
- // CIR: {{%.*}} = cir.cos {{%.*}} : !cir.float
- // LLVM: {{%.*}} = call float @llvm.cos.f32(float {{%.*}})
- // OGCG: {{%.*}} = call float @llvm.cos.f32(float {{%.*}})
- f = __builtin_elementwise_cos(f);
-
- // CIR: {{%.*}} = cir.cos {{%.*}} : !cir.double
- // LLVM: {{%.*}} = call double @llvm.cos.f64(double {{%.*}})
- // OGCG: {{%.*}} = call double @llvm.cos.f64(double {{%.*}})
- d = __builtin_elementwise_cos(d);
-
- // CIR: {{%.*}} = cir.cos {{%.*}} : !cir.vector<4 x !cir.float>
- // LLVM: {{%.*}} = call <4 x float> @llvm.cos.v4f32(<4 x float> {{%.*}})
- // OGCG: {{%.*}} = call <4 x float> @llvm.cos.v4f32(<4 x float> {{%.*}})
- vf4 = __builtin_elementwise_cos(vf4);
-
- // CIR: {{%.*}} = cir.cos {{%.*}} : !cir.vector<4 x !cir.double>
- // LLVM: {{%.*}} = call <4 x double> @llvm.cos.v4f64(<4 x double> {{%.*}})
- // OGCG: {{%.*}} = call <4 x double> @llvm.cos.v4f64(<4 x double> {{%.*}})
- vd4 = __builtin_elementwise_cos(vd4);
-}
-
-void test_builtin_elementwise_floor(float f, double d, vfloat4 vf4,
- vdouble4 vd4) {
- // CIR-LABEL: test_builtin_elementwise_floor
- // LLVM-LABEL: test_builtin_elementwise_floor
- // OGCG-LABEL: test_builtin_elementwise_floor
- // CIR: {{%.*}} = cir.floor {{%.*}} : !cir.float
- // LLVM: {{%.*}} = call float @llvm.floor.f32(float {{%.*}})
- // OGCG: {{%.*}} = call float @llvm.floor.f32(float {{%.*}})
- f = __builtin_elementwise_floor(f);
-
- // CIR: {{%.*}} = cir.floor {{%.*}} : !cir.double
- // LLVM: {{%.*}} = call double @llvm.floor.f64(double {{%.*}})
- // OGCG: {{%.*}} = call double @llvm.floor.f64(double {{%.*}})
- d = __builtin_elementwise_floor(d);
-
- // CIR: {{%.*}} = cir.floor {{%.*}} : !cir.vector<4 x !cir.float>
- // LLVM: {{%.*}} = call <4 x float> @llvm.floor.v4f32(<4 x float> {{%.*}})
- // OGCG: {{%.*}} = call <4 x float> @llvm.floor.v4f32(<4 x float> {{%.*}})
- vf4 = __builtin_elementwise_floor(vf4);
-
- // CIR: {{%.*}} = cir.floor {{%.*}} : !cir.vector<4 x !cir.double>
- // LLVM: {{%.*}} = call <4 x double> @llvm.floor.v4f64(<4 x double> {{%.*}})
- // OGCG: {{%.*}} = call <4 x double> @llvm.floor.v4f64(<4 x double> {{%.*}})
- vd4 = __builtin_elementwise_floor(vd4);
-}
-
-void test_builtin_elementwise_round(float f, double d, vfloat4 vf4,
- vdouble4 vd4) {
- // CIR-LABEL: test_builtin_elementwise_round
- // LLVM-LABEL: test_builtin_elementwise_round
- // OGCG-LABEL: test_builtin_elementwise_round
- // CIR: {{%.*}} = cir.round {{%.*}} : !cir.float
- // LLVM: {{%.*}} = call float @llvm.round.f32(float {{%.*}})
- // OGCG: {{%.*}} = call float @llvm.round.f32(float {{%.*}})
- f = __builtin_elementwise_round(f);
-
- // CIR: {{%.*}} = cir.round {{%.*}} : !cir.double
- // LLVM: {{%.*}} = call double @llvm.round.f64(double {{%.*}})
- // OGCG: {{%.*}} = call double @llvm.round.f64(double {{%.*}})
- d = __builtin_elementwise_round(d);
-
- // CIR: {{%.*}} = cir.round {{%.*}} : !cir.vector<4 x !cir.float>
- // LLVM: {{%.*}} = call <4 x float> @llvm.round.v4f32(<4 x float> {{%.*}})
- // OGCG: {{%.*}} = call <4 x float> @llvm.round.v4f32(<4 x float> {{%.*}})
- vf4 = __builtin_elementwise_round(vf4);
-
- // CIR: {{%.*}} = cir.round {{%.*}} : !cir.vector<4 x !cir.double>
- // LLVM: {{%.*}} = call <4 x double> @llvm.round.v4f64(<4 x double> {{%.*}})
- // OGCG: {{%.*}} = call <4 x double> @llvm.round.v4f64(<4 x double> {{%.*}})
- vd4 = __builtin_elementwise_round(vd4);
-}
-
-void test_builtin_elementwise_rint(float f, double d, vfloat4 vf4,
- vdouble4 vd4) {
- // CIR-LABEL: test_builtin_elementwise_rint
- // LLVM-LABEL: test_builtin_elementwise_rint
- // OGCG-LABEL: test_builtin_elementwise_rint
- // CIR: {{%.*}} = cir.rint {{%.*}} : !cir.float
- // LLVM: {{%.*}} = call float @llvm.rint.f32(float {{%.*}})
- // OGCG: {{%.*}} = call float @llvm.rint.f32(float {{%.*}})
- f = __builtin_elementwise_rint(f);
-
- // CIR: {{%.*}} = cir.rint {{%.*}} : !cir.double
- // LLVM: {{%.*}} = call double @llvm.rint.f64(double {{%.*}})
- // OGCG: {{%.*}} = call double @llvm.rint.f64(double {{%.*}})
- d = __builtin_elementwise_rint(d);
-
- // CIR: {{%.*}} = cir.rint {{%.*}} : !cir.vector<4 x !cir.float>
- // LLVM: {{%.*}} = call <4 x float> @llvm.rint.v4f32(<4 x float> {{%.*}})
- // OGCG: {{%.*}} = call <4 x float> @llvm.rint.v4f32(<4 x float> {{%.*}})
- vf4 = __builtin_elementwise_rint(vf4);
-
- // CIR: {{%.*}} = cir.rint {{%.*}} : !cir.vector<4 x !cir.double>
- // LLVM: {{%.*}} = call <4 x double> @llvm.rint.v4f64(<4 x double> {{%.*}})
- // OGCG: {{%.*}} = call <4 x double> @llvm.rint.v4f64(<4 x double> {{%.*}})
- vd4 = __builtin_elementwise_rint(vd4);
-}
-
-void test_builtin_elementwise_nearbyint(float f, double d, vfloat4 vf4,
- vdouble4 vd4) {
- // CIR-LABEL: test_builtin_elementwise_nearbyint
- // LLVM-LABEL: test_builtin_elementwise_nearbyint
- // OGCG-LABEL: test_builtin_elementwise_nearbyint
- // CIR: {{%.*}} = cir.nearbyint {{%.*}} : !cir.float
- // LLVM: {{%.*}} = call float @llvm.nearbyint.f32(float {{%.*}})
- // OGCG: {{%.*}} = call float @llvm.nearbyint.f32(float {{%.*}})
- f = __builtin_elementwise_nearbyint(f);
-
- // CIR: {{%.*}} = cir.nearbyint {{%.*}} : !cir.double
- // LLVM: {{%.*}} = call double @llvm.nearbyint.f64(double {{%.*}})
- // OGCG: {{%.*}} = call double @llvm.nearbyint.f64(double {{%.*}})
- d = __builtin_elementwise_nearbyint(d);
-
- // CIR: {{%.*}} = cir.nearbyint {{%.*}} : !cir.vector<4 x !cir.float>
- // LLVM: {{%.*}} = call <4 x float> @llvm.nearbyint.v4f32(<4 x float> {{%.*}})
- // OGCG: {{%.*}} = call <4 x float> @llvm.nearbyint.v4f32(<4 x float> {{%.*}})
- vf4 = __builtin_elementwise_nearbyint(vf4);
-
- // CIR: {{%.*}} = cir.nearbyint {{%.*}} : !cir.vector<4 x !cir.double>
- // LLVM: {{%.*}} = call <4 x double> @llvm.nearbyint.v4f64(<4 x double> {{%.*}})
- // OGCG: {{%.*}} = call <4 x double> @llvm.nearbyint.v4f64(<4 x double> {{%.*}})
- vd4 = __builtin_elementwise_nearbyint(vd4);
-}
-
-void test_builtin_elementwise_sin(float f, double d, vfloat4 vf4,
- vdouble4 vd4) {
- // CIR-LABEL: test_builtin_elementwise_sin
- // LLVM-LABEL: test_builtin_elementwise_sin
- // OGCG-LABEL: test_builtin_elementwise...
[truncated]
|
Contributor
Author
|
I was wondering if and where to move the tests |
ayokunle321
changed the title
Contributor
As a general rule, I think it's best to keep things in the equivalent files to those where they appear in the OG tests. Hopefully, we'll be combining them at some point in the distant future. For now, I wouldn't do anything too disruptive to make that happen. |
andykaylor
reviewed
Feb 26, 2026
Contributor
Author
|
@andykaylor addressed comment. |
|
LLVM Buildbot has detected a new failure on builder Full details are available at: https://lab.llvm.org/buildbot/#/builders/140/builds/39545 Here is the relevant piece of the build log for the reference |
|
LLVM Buildbot has detected a new failure on builder Full details are available at: https://lab.llvm.org/buildbot/#/builders/10/builds/23567 Here is the relevant piece of the build log for the reference |
|
LLVM Buildbot has detected a new failure on builder Full details are available at: https://lab.llvm.org/buildbot/#/builders/73/builds/19303 Here is the relevant piece of the build log for the reference |
|
LLVM Buildbot has detected a new failure on builder Full details are available at: https://lab.llvm.org/buildbot/#/builders/206/builds/14639 Here is the relevant piece of the build log for the reference |
|
LLVM Buildbot has detected a new failure on builder Full details are available at: https://lab.llvm.org/buildbot/#/builders/203/builds/38818 Here is the relevant piece of the build log for the reference |
|
LLVM Buildbot has detected a new failure on builder Full details are available at: https://lab.llvm.org/buildbot/#/builders/204/builds/37624 Here is the relevant piece of the build log for the reference |
|
LLVM Buildbot has detected a new failure on builder Full details are available at: https://lab.llvm.org/buildbot/#/builders/205/builds/37603 Here is the relevant piece of the build log for the reference |
sahas3
pushed a commit
to sahas3/llvm-project
that referenced
this pull request
Mar 4, 2026
…#183607) This moves a few tests that were created in the wrong location. Also changes the names of some test files to maintain consistency.
sujianIBM
pushed a commit
to sujianIBM/llvm-project
that referenced
this pull request
Mar 5, 2026
…#183607) This moves a few tests that were created in the wrong location. Also changes the names of some test files to maintain consistency.
This was referenced Apr 15, 2026
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
4 participants
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
This moves a few tests that were created in the wrong location. Also changes the names of some test files to maintain consistency.