Connect CLMatMul function to quantized kernels and resolve NE BatchMatMul int_8 failures
* Adapt the CLMatMul function and ClMatMul operator to use quantized kernels.
* Add function-level tests.
Resolves: COMPMID-5929 and COMPMID-5811
Change-Id: I5348cdcf07b8074c138e04dfef0a73399377accd
Signed-off-by: Jakub Sujak <jakub.sujak@arm.com>
Signed-off-by: Omar Al Khatib <omar.alkhatib@arm.com>
Reviewed-on: https://review.mlplatform.org/c/ml/ComputeLibrary/+/9575
Reviewed-by: Mohmun02 <MohammedSuhail.Munshi@arm.com>
Reviewed-by: SiCong Li <sicong.li@arm.com>
Comments-Addressed: Arm Jenkins <bsgcomp@arm.com>
Tested-by: Arm Jenkins <bsgcomp@arm.com>
diff --git a/src/gpu/cl/kernels/ClMatMulNativeKernel.h b/src/gpu/cl/kernels/ClMatMulNativeKernel.h
index 50aa3b7..f706256 100644
--- a/src/gpu/cl/kernels/ClMatMulNativeKernel.h
+++ b/src/gpu/cl/kernels/ClMatMulNativeKernel.h
@@ -47,17 +47,17 @@
* Dimensions above 2 are collapsed onto dimension 2 and represent the batch.
* @param[in] rhs Input tensor for the RHS matrix. Data type supported: same as @p lhs.
* Dimensions above 2 are collapsed onto dimension 2 and represent the batch.
- * @param[out] output Output tensor info. Data type supported: same as @p lhs
+ * @param[out] dst Output tensor info. Data type supported: same as @p lhs
* @param[in] matmul_info Attributes for Batch MatMul Kernel
*/
- void configure(const ClCompileContext &compile_context, ITensorInfo *lhs, ITensorInfo *rhs, ITensorInfo *output, const MatMulKernelInfo &matmul_info);
+ void configure(const ClCompileContext &compile_context, ITensorInfo *lhs, ITensorInfo *rhs, ITensorInfo *dst, const MatMulKernelInfo &matmul_info);
/** Static function to check if given info will lead to a valid configuration
*
* Similar to @ref ClMatMulNativeKernel::configure()
*
* @return a status
*/
- static Status validate(const ITensorInfo *lhs, const ITensorInfo *rhs, const ITensorInfo *output, const MatMulKernelInfo &matmul_info);
+ static Status validate(const ITensorInfo *lhs, const ITensorInfo *rhs, const ITensorInfo *dst, const MatMulKernelInfo &matmul_info);
// Inherited methods overridden:
void run_op(ITensorPack &tensors, const Window &window, cl::CommandQueue &queue) override;