COMPMID-1995: Update RSH GEMM assembly kernels.

-Updates u8/s8 hybrid dot product kernels to work for any N and any K >=16.
-Adds hybrid FP32 kernels with generic and A55 variants.
-Adds SVE native kernels for fp16/u8/s8.

Change-Id: Ifc0eaba9e3c8ea5bb19d334e870e1b39e4e7e728
Signed-off-by: Georgios Pinitas <georgios.pinitas@arm.com>
Reviewed-on: https://review.mlplatform.org/c/863
Tested-by: Arm Jenkins <bsgcomp@arm.com>
Reviewed-by: Michele Di Giorgio <michele.digiorgio@arm.com>
Comments-Addressed: Arm Jenkins <bsgcomp@arm.com>
Reviewed-by: Michalis Spyrou <michalis.spyrou@arm.com>

1 files changed, 7 insertions, 3 deletions

diff --git a/src/core/NEON/kernels/arm_gemm/gemm_implementation.hpp b/src/core/NEON/kernels/arm_gemm/gemm_implementation.hpp
index bf80784b79..d952140959 100644
--- a/src/core/NEON/kernels/arm_gemm/gemm_implementation.hpp
+++ b/src/core/NEON/kernels/arm_gemm/gemm_implementation.hpp
@@ -112,8 +112,12 @@ bool find_implementation(const GemmArgs<Tret> &args, const GemmImplementation<To
 }
 
 template<typename Top, typename Tret>
-std::vector<std::string> get_compatible_kernels(const GemmArgs<Tret> &args) {
-    std::vector<std::string> res;
+std::vector<KernelDescription> get_compatible_kernels(const GemmArgs<Tret> &args) {
+    std::vector<KernelDescription> res;
+
+    /* Find out what the default implementation in so we can set the flag accordingly later. */
+    const GemmImplementation<Top, Tret> *default_impl;
+    find_implementation(args, default_impl);
 
     auto gemms = gemm_implementation_list<Top, Tret>();
 
@@ -123,7 +127,7 @@ std::vector<std::string> get_compatible_kernels(const GemmArgs<Tret> &args) {
             continue;
         }
 
-        res.push_back(i->name);
+        res.push_back(KernelDescription(i->method, i->name, i==default_impl));
     }
 
     return res;