a43eb4b43e
Keeping the cache misses as low as possible is always a good approach. The opal_list_t is widely used, it should be a highly optimized class. The same functionality can be reached with one one sentinel instead of 2 currently used. I don't have anything against the STL version, but so far nothing can compare with the Knuth algorithm. I replace the current implementation with a modified version of the Knuth algorithm (the one described in The Art of Computer Programming). As expected, the latency went down. This commit was SVN r10776. |
||
---|---|---|
.. | ||
Makefile.am | ||
ompi_bitmap.c | ||
ompi_bitmap.h | ||
ompi_circular_buffer_fifo.h | ||
ompi_fifo.h | ||
ompi_free_list.c | ||
ompi_free_list.h | ||
ompi_pointer_array.c | ||
ompi_pointer_array.h | ||
ompi_rb_tree.c | ||
ompi_rb_tree.h | ||
ompi_seq_tracker.c | ||
ompi_seq_tracker.h |