COMPMID-1021: CPUInfo refactoring.
Removed CPUTarget in favor of the CPUModel type.
CPUInfo now holds a vector of N CPUs.
CPUInfo autoinitialise upon construction with 1 GENERIC CPU.
CPPScheduler fills CPUInfo's vector upon construction (runtime).
IScheduler has a single CPUInfo obj and ThreadInfo always gets a pointer to it (avoid copying the vector)
Change-Id: I30f293258c959c87f6bac5eac8b963beb6a4d365
Reviewed-on: https://eu-gerrit-1.euhpc.arm.com/124626
Tested-by: Jenkins <bsgcomp@arm.com>
Reviewed-by: Anthony Barbier <anthony.barbier@arm.com>
diff --git a/src/runtime/NEON/functions/NEGEMM.cpp b/src/runtime/NEON/functions/NEGEMM.cpp
index c8cba8a..18e6e91 100644
--- a/src/runtime/NEON/functions/NEGEMM.cpp
+++ b/src/runtime/NEON/functions/NEGEMM.cpp
@@ -65,7 +65,8 @@
// Check if we need to reshape the matrix B only on the first run
_reshape_b_only_on_first_run = gemm_info.reshape_b_only_on_first_run();
_run_vector_matrix_multiplication = a->info()->dimension(1) < 2;
- const bool run_optimised = setup_assembly_kernel(a, b, c, d, alpha, beta, _workspace, _memory_group, _asm_glue);
+
+ const bool run_optimised = a->info()->data_type() == DataType::F32 && (c == nullptr || beta == 0.f) && setup_assembly_kernel(a, b, d, alpha, beta, _workspace, _memory_group, _asm_glue);
// Check if the first input tensor is a vector.
// If so, all the kernels for reshaping the tensors can be skipped