diff options
| author | Adnan AlSinan <adnan.alsinan@arm.com> | 2023-10-24 12:03:21 +0100 |
|---|---|---|
| committer | Adnan AlSinan <adnan.alsinan@arm.com> | 2023-10-31 11:00:45 +0000 |
| commit | fde45d836cf753a94915ac42d8a13da7edc52221 (patch) | |
| tree | 6ed787749aa3caec13a0b3c2c64ea591b423089c /src/dynamic_fusion/sketch/gpu/operators/GpuMatMul.cpp | |
| parent | 5ef0bdd53dd2ce6bc7ad28077ffac3bf9e939b5f (diff) | |
| download | ComputeLibrary-fde45d836cf753a94915ac42d8a13da7edc52221.tar.gz | |
Extend CKW MatMul with nt_t
- Add the kernel variant: (nt_t) to GpuCKWMatMul.
- Extend CKW MatMul validation test with nt_t.
- Fixes a bug in CKW where z-dim = 1.
Resolves: COMPMID-6435
Signed-off-by: Adnan AlSinan <adnan.alsinan@arm.com>
Change-Id: I4c5e8791e55f21ffff3c11eca7802c51a4259977
Reviewed-on: https://review.mlplatform.org/c/ml/ComputeLibrary/+/10525
Comments-Addressed: Arm Jenkins <bsgcomp@arm.com>
Tested-by: Arm Jenkins <bsgcomp@arm.com>
Reviewed-by: Gian Marco Iodice <gianmarco.iodice@arm.com>
Benchmark: Arm Jenkins <bsgcomp@arm.com>
Diffstat (limited to 'src/dynamic_fusion/sketch/gpu/operators/GpuMatMul.cpp')
| -rw-r--r-- | src/dynamic_fusion/sketch/gpu/operators/GpuMatMul.cpp | 2 |
1 files changed, 0 insertions, 2 deletions
diff --git a/src/dynamic_fusion/sketch/gpu/operators/GpuMatMul.cpp b/src/dynamic_fusion/sketch/gpu/operators/GpuMatMul.cpp index ee27b5ea47..e24629a036 100644 --- a/src/dynamic_fusion/sketch/gpu/operators/GpuMatMul.cpp +++ b/src/dynamic_fusion/sketch/gpu/operators/GpuMatMul.cpp @@ -87,8 +87,6 @@ Status is_supported_op_helper(const GpuWorkloadContext &context, // Check support level // Data type ARM_COMPUTE_RETURN_ERROR_ON_DATA_TYPE_CHANNEL_NOT_IN(lhs, 1, DataType::F16, DataType::F32); - // Data layout - ARM_COMPUTE_RETURN_ERROR_ON_DATA_LAYOUT_NOT_IN(lhs, DataLayout::NHWC); // Check components if (context.gpu_language() == GpuLanguage::OpenCL) |