diff options
author | Sheri Zhang <sheri.zhang@arm.com> | 2021-05-06 10:05:22 +0100 |
---|---|---|
committer | Sheri Zhang <sheri.zhang@arm.com> | 2021-05-07 09:58:48 +0000 |
commit | 78e76912f0758c60e83922098231645efc02d8d6 (patch) | |
tree | b40f3cbc159be96c8758e908613e069eba4a12bd /src/runtime | |
parent | 3ae6580d26cee273b750e69b5c06f6efa4caf3fb (diff) | |
download | ComputeLibrary-78e76912f0758c60e83922098231645efc02d8d6.tar.gz |
Remove TODOs
Remove TODO/FIXME either already done or won't do.
Partially resolve: COMPMID-4471
Signed-off-by: Sheri Zhang <sheri.zhang@arm.com>
Change-Id: Iec8f25b9bf1b2a70c072dd17d44625fa93e84ed1
Reviewed-on: https://review.mlplatform.org/c/ml/ComputeLibrary/+/5591
Reviewed-by: Georgios Pinitas <georgios.pinitas@arm.com>
Comments-Addressed: Arm Jenkins <bsgcomp@arm.com>
Tested-by: Georgios Pinitas <georgios.pinitas@arm.com>
Diffstat (limited to 'src/runtime')
-rw-r--r-- | src/runtime/NEON/functions/NEGEMMAssemblyDispatch.cpp | 2 | ||||
-rw-r--r-- | src/runtime/cpu/operators/CpuDepthwiseConvolutionAssemblyDispatch.cpp | 1 |
2 files changed, 1 insertions, 2 deletions
diff --git a/src/runtime/NEON/functions/NEGEMMAssemblyDispatch.cpp b/src/runtime/NEON/functions/NEGEMMAssemblyDispatch.cpp index 7fb1d583ff..c58a662f10 100644 --- a/src/runtime/NEON/functions/NEGEMMAssemblyDispatch.cpp +++ b/src/runtime/NEON/functions/NEGEMMAssemblyDispatch.cpp @@ -75,7 +75,7 @@ Params extract_parameters(const ITensor *a, const ITensor *b, const ITensor *d, else { p.multis = b->info()->tensor_shape().z(); - p.batches = d->info()->tensor_shape().total_size_upper(2) / p.multis; //COMPMID-1423: Agree on and document the layout of gemm inputs/outputs + p.batches = d->info()->tensor_shape().total_size_upper(2) / p.multis; } // Update M in case of GEMM3D for output diff --git a/src/runtime/cpu/operators/CpuDepthwiseConvolutionAssemblyDispatch.cpp b/src/runtime/cpu/operators/CpuDepthwiseConvolutionAssemblyDispatch.cpp index 5f5304cded..039714abb1 100644 --- a/src/runtime/cpu/operators/CpuDepthwiseConvolutionAssemblyDispatch.cpp +++ b/src/runtime/cpu/operators/CpuDepthwiseConvolutionAssemblyDispatch.cpp @@ -459,7 +459,6 @@ bool CpuDepthwiseConvolutionAssemblyDispatch::is_optimized_supported(const ITens } // Check data type - // TODO (COMPMID-3004): Add assembly optimized routine for QASYMM8_SIGNED NEDepthwiseConvolutionLayer const DataType input_type = input->data_type(); const bool is_input_type_valid = is_data_type_float(input_type) || input_type == DataType::QASYMM8; const DataType weights_type = weights->data_type(); |