diff options
author | Michele Di Giorgio <michele.digiorgio@arm.com> | 2020-07-06 11:27:21 +0100 |
---|---|---|
committer | Georgios Pinitas <georgios.pinitas@arm.com> | 2020-07-22 10:35:29 +0000 |
commit | f932d2c8409831cb9cb97a2eb65be93ad4709cd6 (patch) | |
tree | 44fd816205d9b908deaff9f21e5018da42e3cd80 /src/runtime/NEON | |
parent | f20d6d6ae5a0da2c856294e93341cdc065db58f9 (diff) | |
download | ComputeLibrary-f932d2c8409831cb9cb97a2eb65be93ad4709cd6.tar.gz |
COMPMID-3386: Support memory injection in CLConcatenate functions/kernels
Signed-off-by: Georgios Pinitas <georgios.pinitas@arm.com>
Change-Id: I611adf4f506d406540e920b0bd6befb4b5108918
Reviewed-on: https://review.mlplatform.org/c/ml/ComputeLibrary/+/3601
Tested-by: Arm Jenkins <bsgcomp@arm.com>
Reviewed-by: Michalis Spyrou <michalis.spyrou@arm.com>
Comments-Addressed: Arm Jenkins <bsgcomp@arm.com>
Diffstat (limited to 'src/runtime/NEON')
-rw-r--r-- | src/runtime/NEON/functions/NEGEMMAssemblyDispatch.cpp | 3 |
1 files changed, 2 insertions, 1 deletions
diff --git a/src/runtime/NEON/functions/NEGEMMAssemblyDispatch.cpp b/src/runtime/NEON/functions/NEGEMMAssemblyDispatch.cpp index 8e9f393da5..3b9dde2bf7 100644 --- a/src/runtime/NEON/functions/NEGEMMAssemblyDispatch.cpp +++ b/src/runtime/NEON/functions/NEGEMMAssemblyDispatch.cpp @@ -438,7 +438,8 @@ void Fallback<TypeInput, TypeOutput, OutputStage>::run() const int granule_threshold = 200; scheduling_hint = IScheduler::Hints(Window::DimX, IScheduler::StrategyHint::DYNAMIC, granule_threshold); } - else if(_kernel_info.method == arm_gemm::GemmMethod::GEMM_INTERLEAVED_2D && (_d->info()->data_type() == DataType::F32 || _d->info()->data_type() == DataType::F16 || _d->info()->data_type() == DataType::U8 || _d->info()->data_type() == DataType::S8) ) + else if(_kernel_info.method == arm_gemm::GemmMethod::GEMM_INTERLEAVED_2D && (_d->info()->data_type() == DataType::F32 || _d->info()->data_type() == DataType::F16 + || _d->info()->data_type() == DataType::U8 || _d->info()->data_type() == DataType::S8)) { //GEMM_INTERLEAVED supports 2D parallelism, IScheduler::split_dimensions_all signals to parallelise over all window dimensions const int granule_threshold = 200; |