| commit | 728d3cf79fb268c3069c31840271ae08b3993842 | [log] [tgz] |
|---|---|---|
| author | Georgios Pinitas <georgios.pinitas@arm.com> | Fri Sep 21 13:41:35 2018 +0100 |
| committer | Anthony Barbier <anthony.barbier@arm.com> | Fri Nov 02 16:54:54 2018 +0000 |
| tree | d23254a997918d5d8ca0b8bd26d50ed5e8b91091 | |
| parent | a6825a427a51da815a805d66ce65c98de282d89b [diff] |
COMPMID-1598 : Fix compilation error in CLDepthwiseConvolutionQS8 kernel Change-Id: I65eeb0cba2af462c6ef64a536ad263c407d62811 Reviewed-on: https://eu-gerrit-1.euhpc.arm.com/149609 Reviewed-by: Michalis Spyrou <michalis.spyrou@arm.com> Tested-by: bsgcomp <bsgcomp@arm.com>
diff --git a/src/core/CL/cl_kernels/depthwise_convolution_quantized.cl b/src/core/CL/cl_kernels/depthwise_convolution_quantized.cl index b3edc52..7cd4879 100644 --- a/src/core/CL/cl_kernels/depthwise_convolution_quantized.cl +++ b/src/core/CL/cl_kernels/depthwise_convolution_quantized.cl
@@ -138,7 +138,7 @@ #if defined(HAS_BIAS) Vector biases = CONVERT_TO_VECTOR_STRUCT_NO_STEP(biases); - int bias_value = *((__global int *)(vector_offset(&biases, channel)); + int bias_value = *((__global int *)(vector_offset(&biases, channel))); #endif //defined(HAS_BIAS) // Load relevant input and weights data (Accounts depth multiplier when indexing input, OFM = IFM * DEPTH_MULTIPLIER)