diff options
| author | Viet-Hoa Do <viet-hoa.do@arm.com> | 2024-06-14 15:53:40 +0100 |
|---|---|---|
| committer | Viet-Hoa Do <viet-hoa.do@arm.com> | 2024-06-14 16:59:48 +0000 |
| commit | cbc27ff3ebd93f9939a6c2248626b87898545069 (patch) | |
| tree | 6ac8887cbf11d8eadd44d65595acf9a764609b39 /src/core/cpu/kernels/pooling/neon | |
| parent | e8f28136a048829211d30208cdd6679f0f1c1632 (diff) | |
| download | ComputeLibrary-main.tar.gz | |
* Non-optimized FP16 GeMM kernel has out-of-bound memory write.
- This doesn't affect optimized assembly kernels.
- This bug writes 1 extra FP16 value to the destination tensor.
Resolves: COMPMID-6904
Signed-off-by: Viet-Hoa Do <viet-hoa.do@arm.com>
Change-Id: I26b8ebcd15680b25c97c4b7e331996f397692447
Reviewed-on: https://review.mlplatform.org/c/ml/ComputeLibrary/+/11706
Reviewed-by: Jakub Sujak <jakub.sujak@arm.com>
Comments-Addressed: Arm Jenkins <bsgcomp@arm.com>
Benchmark: Arm Jenkins <bsgcomp@arm.com>
Tested-by: Arm Jenkins <bsgcomp@arm.com>
Diffstat (limited to 'src/core/cpu/kernels/pooling/neon')
0 files changed, 0 insertions, 0 deletions