diff options
| author | Michele Di Giorgio <michele.digiorgio@arm.com> | 2020-07-28 16:59:48 +0100 |
|---|---|---|
| committer | Michele Di Giorgio <michele.digiorgio@arm.com> | 2020-07-29 13:10:05 +0000 |
| commit | a179798501af2b3939f9282b2f03ef4f98471d81 (patch) | |
| tree | eb437c20ca1fc33641534fc81f36d1296acb1b67 /src/core/CL/cl_kernels/winograd_output_transform.cl | |
| parent | 42e57362933b7fa04f9a86297fa75ed38c5e3c97 (diff) | |
| download | ComputeLibrary-a179798501af2b3939f9282b2f03ef4f98471d81.tar.gz | |
COMPMID-3585: Android R while_fib_n_5_quant8 failure on CpuAcc
Assembly kernels do not support quantized GEMM when the multiplier is
greater than 1 (which leads to negative result shift used for
requantization).
Change-Id: I7a766cd0d13f549d217613ca67bc952923f309de
Signed-off-by: Michele Di Giorgio <michele.digiorgio@arm.com>
Reviewed-on: https://review.mlplatform.org/c/ml/ComputeLibrary/+/3538
Tested-by: Arm Jenkins <bsgcomp@arm.com>
Reviewed-by: Georgios Pinitas <georgios.pinitas@arm.com>
Comments-Addressed: Arm Jenkins <bsgcomp@arm.com>
Diffstat (limited to 'src/core/CL/cl_kernels/winograd_output_transform.cl')
0 files changed, 0 insertions, 0 deletions