diff options
author | Gian Marco Iodice <gianmarco.iodice@arm.com> | 2021-07-07 10:25:41 +0100 |
---|---|---|
committer | Gian Marco Iodice <gianmarco.iodice@arm.com> | 2021-07-09 14:17:39 +0000 |
commit | 04b39e8e56112dabf6f5746117354680a9985841 (patch) | |
tree | af603999edcf0c82ac1fd6a861e3f8335befedf5 /src/runtime/NEON/functions/NEWinogradConvolutionLayer.cpp | |
parent | 24b892072a2bd8190ba63d09fb0082113d7d032a (diff) | |
download | ComputeLibrary-04b39e8e56112dabf6f5746117354680a9985841.tar.gz |
Limit the LOOP_UNROLLING on the kernel height
To reduce the risk of having a long OpenCL kernel, we limit the loop
unrolling on the kernel height. In particular, we unroll only if the
kernel height is less than or equal to 5
Resolves COMPMID-4604
Change-Id: Iece787989f36afb90f1c7676b53d9015e652bdbd
Signed-off-by: Gian Marco Iodice <gianmarco.iodice@arm.com>
Reviewed-on: https://review.mlplatform.org/c/ml/ComputeLibrary/+/5916
Tested-by: Arm Jenkins <bsgcomp@arm.com>
Reviewed-by: Michele Di Giorgio <michele.digiorgio@arm.com>
Reviewed-by: Giorgio Arena <giorgio.arena@arm.com>
Comments-Addressed: Arm Jenkins <bsgcomp@arm.com>
Diffstat (limited to 'src/runtime/NEON/functions/NEWinogradConvolutionLayer.cpp')
0 files changed, 0 insertions, 0 deletions