diff options
| author | Georgios Pinitas <georgios.pinitas@arm.com> | 2020-11-11 21:05:24 +0000 |
|---|---|---|
| committer | Sheri Zhang <sheri.zhang@arm.com> | 2020-11-13 10:34:52 +0000 |
| commit | 412b789582c5992431028e9b91c4d8b99d5c4900 (patch) | |
| tree | 62f7750c113dae21c877f6ed384b73c24dc4221b /src/core/NEON/kernels/NEDepthwiseConvolutionLayerNativeKernel.cpp | |
| parent | d7341fb9e3b24b904edf7ac9d83e1e063bc77765 (diff) | |
| download | ComputeLibrary-412b789582c5992431028e9b91c4d8b99d5c4900.tar.gz | |
COMPMID-3852: Fix NEReduction window
ReductionOperations splits the kernel for scheduling on the X
dimension when reduction axis is > 0.
By setting the execution window to be unit one in the X dimension the
execution was always restricted to a single thread.
Alters the window to enable multi-threading
Signed-off-by: Georgios Pinitas <georgios.pinitas@arm.com>
Change-Id: Idcbe2b78957678310bb8e895969f01de972d3667
Reviewed-on: https://review.mlplatform.org/c/ml/ComputeLibrary/+/4389
Comments-Addressed: Arm Jenkins <bsgcomp@arm.com>
Tested-by: Arm Jenkins <bsgcomp@arm.com>
Reviewed-by: Michele Di Giorgio <michele.digiorgio@arm.com>
Diffstat (limited to 'src/core/NEON/kernels/NEDepthwiseConvolutionLayerNativeKernel.cpp')
0 files changed, 0 insertions, 0 deletions