Actionable insights tagged with ‘rmk’ in Parallelware Analyzer:
RMK001: loop nesting that might benefit from hybrid parallelization using multithreading and SIMD
RMK002: Loop nesting that might benefit from hybrid parallelization using offloading and SIMD
RMK003: Potential temporary variable for the loop which might be privatizable, thus enabling the loop parallelization
RMK004: Avoid strided array access to improve performance
RMK005: Avoid non-consecutive array access to improve performance
RMK006: Avoid indirect array access to improve performance
