Related to the OpenMP standard
Actionable insights tagged with ‘omp’ in Parallelware Analyzer:
PWD002: Unprotected multithreading reduction operation
PWD003: Missing array range in data copy to the GPU
PWD007: Unprotected multithreading recurrence
PWD008: Unprotected multithreading recurrence due to out-of-dimension-bounds array access
PWD009: Incorrect privatization in OpenMP parallel region
PWD010: Incorrect sharing in OpenMP parallel region
PWD011: Missing OpenMP last private clause
PWR004: Declare OpenMP scoping for all variables
PWR005: Disable default OpenMP scoping
PWR006: Avoid privatization of read-only variables
PWR009: Use OpenMP teams to offload work to GPU
PWR010: Avoid column-major array access in C/C++
PWR013: Avoid copying unused variables to the GPU
PWR015: Avoid copying unnecessary array elements to or from the GPU
PWR024: Loop can be rewritten in OpenMP canonical form
PWR025: Consider annotating pure function with OpenMP ‘declare simd’
