Revert "[llvm] Improve llvm.objectsize computation by computing GEP, alloca and mallo...
[llvm-project.git] / clang / test / CodeGenOpenCL / builtins-amdgcn-dl-insts.cl
blobd5478609851b0f3b7a190561d8142f11c4a6393a
1 // REQUIRES: amdgpu-registered-target
3 // RUN: %clang_cc1 -triple amdgcn-unknown-unknown -target-cpu gfx906 -emit-llvm -o - %s | FileCheck %s
4 // RUN: %clang_cc1 -triple amdgcn-unknown-unknown -target-cpu gfx1011 -emit-llvm -o - %s | FileCheck %s
5 // RUN: %clang_cc1 -triple amdgcn-unknown-unknown -target-cpu gfx1012 -emit-llvm -o - %s | FileCheck %s
7 typedef unsigned int uint;
8 typedef half __attribute__((ext_vector_type(2))) half2;
9 typedef short __attribute__((ext_vector_type(2))) short2;
10 typedef unsigned short __attribute__((ext_vector_type(2))) ushort2;
12 // CHECK-LABEL: @builtins_amdgcn_dl_insts
13 // CHECK: call float @llvm.amdgcn.fdot2(<2 x half> %v2hA, <2 x half> %v2hB, float %fC, i1 false)
14 // CHECK: call float @llvm.amdgcn.fdot2(<2 x half> %v2hA, <2 x half> %v2hB, float %fC, i1 true)
16 // CHECK: call i32 @llvm.amdgcn.sdot2(<2 x i16> %v2ssA, <2 x i16> %v2ssB, i32 %siC, i1 false)
17 // CHECK: call i32 @llvm.amdgcn.sdot2(<2 x i16> %v2ssA, <2 x i16> %v2ssB, i32 %siC, i1 true)
19 // CHECK: call i32 @llvm.amdgcn.udot2(<2 x i16> %v2usA, <2 x i16> %v2usB, i32 %uiC, i1 false)
20 // CHECK: call i32 @llvm.amdgcn.udot2(<2 x i16> %v2usA, <2 x i16> %v2usB, i32 %uiC, i1 true)
22 // CHECK: call i32 @llvm.amdgcn.sdot4(i32 %siA, i32 %siB, i32 %siC, i1 false)
23 // CHECK: call i32 @llvm.amdgcn.sdot4(i32 %siA, i32 %siB, i32 %siC, i1 true)
25 // CHECK: call i32 @llvm.amdgcn.udot4(i32 %uiA, i32 %uiB, i32 %uiC, i1 false)
26 // CHECK: call i32 @llvm.amdgcn.udot4(i32 %uiA, i32 %uiB, i32 %uiC, i1 true)
28 // CHECK: call i32 @llvm.amdgcn.sdot8(i32 %siA, i32 %siB, i32 %siC, i1 false)
29 // CHECK: call i32 @llvm.amdgcn.sdot8(i32 %siA, i32 %siB, i32 %siC, i1 true)
31 // CHECK: call i32 @llvm.amdgcn.udot8(i32 %uiA, i32 %uiB, i32 %uiC, i1 false)
32 // CHECK: call i32 @llvm.amdgcn.udot8(i32 %uiA, i32 %uiB, i32 %uiC, i1 true)
33 kernel void builtins_amdgcn_dl_insts(
34 global float *fOut, global int *siOut, global uint *uiOut,
35 half2 v2hA, half2 v2hB, float fC,
36 short2 v2ssA, short2 v2ssB, int siA, int siB, int siC,
37 ushort2 v2usA, ushort2 v2usB, uint uiA, uint uiB, uint uiC) {
38 fOut[0] = __builtin_amdgcn_fdot2(v2hA, v2hB, fC, false);
39 fOut[1] = __builtin_amdgcn_fdot2(v2hA, v2hB, fC, true);
41 siOut[0] = __builtin_amdgcn_sdot2(v2ssA, v2ssB, siC, false);
42 siOut[1] = __builtin_amdgcn_sdot2(v2ssA, v2ssB, siC, true);
44 uiOut[0] = __builtin_amdgcn_udot2(v2usA, v2usB, uiC, false);
45 uiOut[1] = __builtin_amdgcn_udot2(v2usA, v2usB, uiC, true);
47 siOut[2] = __builtin_amdgcn_sdot4(siA, siB, siC, false);
48 siOut[3] = __builtin_amdgcn_sdot4(siA, siB, siC, true);
50 uiOut[2] = __builtin_amdgcn_udot4(uiA, uiB, uiC, false);
51 uiOut[3] = __builtin_amdgcn_udot4(uiA, uiB, uiC, true);
53 siOut[4] = __builtin_amdgcn_sdot8(siA, siB, siC, false);
54 siOut[5] = __builtin_amdgcn_sdot8(siA, siB, siC, true);
56 uiOut[4] = __builtin_amdgcn_udot8(uiA, uiB, uiC, false);
57 uiOut[5] = __builtin_amdgcn_udot8(uiA, uiB, uiC, true);