derek.holloway
7502bf64a4
CodeQL / Analyze (actions) (push) Canceled after 0s
CodeQL / Analyze (python) (push) Canceled after 0s
pre-commit / pre-commit (push) Canceled after 0s
Multi-Arch CI / setup (push) Canceled after 0s
Publish build_manylinux_x86_64 images / publish_build_manylinux_x86_64 (push) Canceled after 0s
Setup (Multi-Arch) / setup (push) Canceled after 0s
Multi-Arch CI / Linux::${{ fromJSON(needs.setup.outputs.linux_build_config || '{}').build_variant_label || 'skip' }} (push) Canceled after 0s
Publish TheRock Docker image / build-and-push-image (push) Canceled after 0s
Publish no_rocm_image_ubi10 image / publish_no_rocm_image_ubi10 (push) Canceled after 0s
Multi-Arch CI / Windows::${{ fromJSON(needs.setup.outputs.windows_build_config || '{}').build_variant_label || 'skip' }} (push) Canceled after 0s
Publish TheRock Docker image / build-and-push-image (push) Canceled after 0s
Multi-Arch CI / Manifest Diff (push) Canceled after 0s
TheRock Manifest Diff Report / Generate Manifest Diff Report (push) Canceled after 0s
Multi-Arch CI / CI Summary (push) Canceled after 0s
Publish no_rocm_image_ubuntu24_04 images / publish_no_rocm_image_ubuntu24_04 (push) Canceled after 0s
Publish no_rocm_image_ubuntu24_04_ocl_rt images / publish_no_rocm_image_ubuntu24_04_ocl_rt (push) Canceled after 0s
Publish TheRock Docker image / build-and-push-image (push) Canceled after 0s
Publish TheRock Docker image / build-and-push-image (push) Canceled after 0s
Publish no_rocm_image_ubuntu24_04 images / publish_no_rocm_image_ubuntu24_04_rocgdb (push) Canceled after 0s
Publish no_rocm_image_ubuntu24_04_rocgdb images / publish_no_rocm_image_ubuntu24_04_rocgdb (push) Canceled after 0s
Publish no_rocm_image_ubuntu24_04 images / publish_no_rocm_image_ubuntu24_04_ocl_rt (push) Canceled after 0s
Publish TheRock Docker image / build-and-push-image (push) Canceled after 0s
Unit Tests / Unit Tests :: ubuntu-24.04 (push) Canceled after 0s
Unit Tests / Unit Tests :: windows-2022 (push) Canceled after 0s
Publish no_rocm_image_ubuntu24_04 images / publish_no_rocm_image_ubuntu24_04_media (push) Canceled after 0s
Unit Tests / Unit Tests Summary (push) Canceled after 0s
Publish no_rocm_image_ubuntu24_04 images / publish_no_rocm_image_ubuntu24_04_openmpi (push) Canceled after 0s
69 lines
2.5 KiB
Common Lisp
69 lines
2.5 KiB
Common Lisp
// RUN: %clang_cc1 -triple amdgcn-unknown-unknown -target-cpu gfx1100 -target-feature +wavefrontsize32 -DWMMA_GFX1100_TESTS -S -o - %s | FileCheck %s --check-prefix=CHECK-GFX1100
|
|
|
|
typedef float v4f __attribute__((ext_vector_type(4)));
|
|
typedef float v8f __attribute__((ext_vector_type(8)));
|
|
typedef half v16h __attribute__((ext_vector_type(16)));
|
|
typedef int v2i __attribute__((ext_vector_type(2)));
|
|
typedef int v4i __attribute__((ext_vector_type(4)));
|
|
typedef int v8i __attribute__((ext_vector_type(8)));
|
|
typedef short v16s __attribute__((ext_vector_type(16)));
|
|
|
|
#ifdef WMMA_GFX1100_TESTS
|
|
|
|
// Wave32
|
|
|
|
|
|
// CHECK-GFX1100-LABEL: test_amdgcn_wmma_f32_16x16x16_f16_w32:
|
|
// CHECK-GFX1100: v_wmma_f32_16x16x16_f16 v[{{.*}}], v[{{.*}} v[{{.*}}], v[{{.*}}]
|
|
//
|
|
void test_amdgcn_wmma_f32_16x16x16_f16_w32(global v8f* out, v16h a, v16h b, v8f c)
|
|
{
|
|
*out = __builtin_amdgcn_wmma_f32_16x16x16_f16_w32(a, b, c);
|
|
}
|
|
|
|
|
|
// CHECK-GFX1100-LABEL: test_amdgcn_wmma_f32_16x16x16_bf16_w32:
|
|
// CHECK-GFX1100: v_wmma_f32_16x16x16_bf16 v[{{.*}}], v[{{.*}} v[{{.*}}], v[{{.*}}]
|
|
//
|
|
void test_amdgcn_wmma_f32_16x16x16_bf16_w32(global v8f* out, v16s a, v16s b, v8f c)
|
|
{
|
|
*out = __builtin_amdgcn_wmma_f32_16x16x16_bf16_w32(a, b, c);
|
|
}
|
|
|
|
|
|
// CHECK-GFX1100-LABEL: test_amdgcn_wmma_f16_16x16x16_f16_w32:
|
|
// CHECK-GFX1100: v_wmma_f16_16x16x16_f16 v[{{.*}}], v[{{.*}} v[{{.*}}], v[{{.*}}] op_sel:[0,0,1]
|
|
//
|
|
void test_amdgcn_wmma_f16_16x16x16_f16_w32(global v16h* out, v16h a, v16h b, v16h c)
|
|
{
|
|
*out = __builtin_amdgcn_wmma_f16_16x16x16_f16_w32(a, b, c, true);
|
|
}
|
|
|
|
|
|
// CHECK-GFX1100-LABEL: test_amdgcn_wmma_bf16_16x16x16_bf16_w32:
|
|
// CHECK-GFX1100: v_wmma_bf16_16x16x16_bf16 v[{{.*}}], v[{{.*}} v[{{.*}}], v[{{.*}}] op_sel:[0,0,1]
|
|
//
|
|
void test_amdgcn_wmma_bf16_16x16x16_bf16_w32(global v16s* out, v16s a, v16s b, v16s c)
|
|
{
|
|
*out = __builtin_amdgcn_wmma_bf16_16x16x16_bf16_w32(a, b, c, true);
|
|
}
|
|
|
|
|
|
// CHECK-GFX1100-LABEL: test_amdgcn_wmma_i32_16x16x16_iu8_w32:
|
|
// CHECK-GFX1100: v_wmma_i32_16x16x16_iu8 v[{{.*}}], v[{{.*}} v[{{.*}} v[{{.*}}] neg_lo:[1,1,0]
|
|
//
|
|
void test_amdgcn_wmma_i32_16x16x16_iu8_w32(global v8i* out, v4i a, v4i b, v8i c)
|
|
{
|
|
*out = __builtin_amdgcn_wmma_i32_16x16x16_iu8_w32(true, a, true, b, c, false);
|
|
}
|
|
|
|
|
|
// CHECK-GFX1100-LABEL: test_amdgcn_wmma_i32_16x16x16_iu4_w32:
|
|
// CHECK-GFX1100: v_wmma_i32_16x16x16_iu4 v[{{.*}}, v[{{.*}} v[{{.*}} v[{{.*}} neg_lo:[1,1,0]
|
|
void test_amdgcn_wmma_i32_16x16x16_iu4_w32(global v8i* out, v2i a, v2i b, v8i c)
|
|
{
|
|
*out = __builtin_amdgcn_wmma_i32_16x16x16_iu4_w32(true, a, true, b, c, false);
|
|
}
|
|
|
|
#endif
|