diff options
| author | Usama Arif <usama.arif@arm.com> | 2019-05-24 11:13:20 +0100 |
|---|---|---|
| committer | Usama Arif <usama.arif@arm.com> | 2019-05-24 13:36:49 +0000 |
| commit | c5c750dc63a7f97ef872cdea9140d17578d21547 (patch) | |
| tree | e09a3b026e2b1405022400cb396cdb4841633514 /src/core/NEON/kernels/NEDepthConvertLayerKernel.cpp | |
| parent | 22f917cfd2282052c447068b188eee0c59f737fd (diff) | |
| download | ComputeLibrary-c5c750dc63a7f97ef872cdea9140d17578d21547.tar.gz | |
COMPMID-2252: Saturate when converting from S32 to U8 IN NEDepthConvertLayerKernel
Change-Id: Ic1f7500cf5d1ab669abf8a5540a8493e3b5ce4d3
Signed-off-by: Usama Arif <usama.arif@arm.com>
Reviewed-on: https://eu-gerrit-1.euhpc.arm.com/186336
Tested-by: bsgcomp <bsgcomp@arm.com>
Reviewed-by: Manuel Bottini <manuel.bottini@arm.com>
Comments-Addressed: bsgcomp <bsgcomp@arm.com>
Reviewed-on: https://review.mlplatform.org/c/1220
Tested-by: Arm Jenkins <bsgcomp@arm.com>
Reviewed-by: Michalis Spyrou <michalis.spyrou@arm.com>
Reviewed-by: Pablo Marquez <pablo.tello@arm.com>
Comments-Addressed: Arm Jenkins <bsgcomp@arm.com>
Diffstat (limited to 'src/core/NEON/kernels/NEDepthConvertLayerKernel.cpp')
| -rw-r--r-- | src/core/NEON/kernels/NEDepthConvertLayerKernel.cpp | 5 |
1 files changed, 2 insertions, 3 deletions
diff --git a/src/core/NEON/kernels/NEDepthConvertLayerKernel.cpp b/src/core/NEON/kernels/NEDepthConvertLayerKernel.cpp index 531873e49e..10bbe8324c 100644 --- a/src/core/NEON/kernels/NEDepthConvertLayerKernel.cpp +++ b/src/core/NEON/kernels/NEDepthConvertLayerKernel.cpp @@ -649,9 +649,8 @@ void NEDepthConvertLayerKernel::run(const Window &window, const ThreadInfo &info vqshlq_s32(vld1q_s32(reinterpret_cast<int32_t *>(input.ptr()) + 12), b) } }; - - vst1_u8(reinterpret_cast<uint8_t *>(output.ptr()), vmovn_u16(vcombine_u16(vqmovun_s32(texels.val[0]), vqmovun_s32(texels.val[1])))); - vst1_u8(reinterpret_cast<uint8_t *>(output.ptr())+8, vmovn_u16(vcombine_u16(vqmovun_s32(texels.val[2]), vqmovun_s32(texels.val[3])))); + vst1_u8(reinterpret_cast<uint8_t *>(output.ptr()), vqmovn_u16(vcombine_u16(vqmovun_s32(texels.val[0]), vqmovun_s32(texels.val[1])))); + vst1_u8(reinterpret_cast<uint8_t *>(output.ptr())+8, vqmovn_u16(vcombine_u16(vqmovun_s32(texels.val[2]), vqmovun_s32(texels.val[3])))); }, input, output); } |