[ONCPUML-951] Variable weight support for Convolution.

API changes for NEGEMMConvolutionLayer and CpuGemmConv2d

Built with:

    scons neon=1 opencl=0 os=linux arch=armv8.2-a multi_isa=1 \
        build=native -j32 Werror=false validation_tests=1 build_dir=opt \
        standalone=1 asserts=1 experimental_fixed_format_kernels=1 .

Tested with:

    ./build/opt/tests/arm_compute_validation

Hardware where the test executable was run:

Neoverse N1

Test coverage:

* NEGEMMConvolutionLayer, CpuGemmConv2d
* NHWC (the only one supported by the fixed-format kernels)
* F16, F32
* Shapes: RunSmall

Change-Id: I4fd3e495a7cbf61210ea02d37440ba9652934e99
Signed-off-by: Francesco Petrogalli <francesco.petrogalli@arm.com>
Reviewed-on: https://review.mlplatform.org/c/ml/ComputeLibrary/+/7632
Tested-by: Arm Jenkins <bsgcomp@arm.com>
Reviewed-by: Gunes Bayir <gunes.bayir@arm.com>
Comments-Addressed: Arm Jenkins <bsgcomp@arm.com>
Benchmark: Arm Jenkins <bsgcomp@arm.com>

1 files changed, 5 insertions, 2 deletions

diff --git a/src/core/NEON/kernels/arm_gemm/gemm_implementation.hpp b/src/core/NEON/kernels/arm_gemm/gemm_implementation.hpp
index 75fb1cb306..19c8fcadd3 100644
--- a/src/core/NEON/kernels/arm_gemm/gemm_implementation.hpp
+++ b/src/core/NEON/kernels/arm_gemm/gemm_implementation.hpp
@@ -306,9 +306,12 @@ std::vector<KernelDescription> get_compatible_kernels(const GemmArgs &args, cons
 }
 
 template<typename Top, typename Tret, class OutputStage>
-bool has_opt_gemm(const GemmArgs &args, const OutputStage &os) {
+bool has_opt_gemm(WeightFormat &wf, const GemmArgs &args, const OutputStage &os) {
     const GemmImplementation<Top, Tret, OutputStage> *impl;
-    return find_implementation<Top, Tret, OutputStage>(args, os, impl);
+    const bool success =  find_implementation<Top, Tret, OutputStage>(args, os, impl);
+    if (success)
+      wf = UniqueGemmCommon<Top, Tret>(impl->do_instantiate(args, os))->get_config().weight_format;
+    return success;
 }
 
 template<typename Top, typename Tret, class OutputStage>