George Bosilca a43eb4b43e It's not about how much memory we use, but about how we use it. ... 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.		2006-07-13 04:56:15 +00:00
..
Makefile.am	Seperate out sequence tracking list as stand alone class.	2006-03-23 17:02:17 +00:00
ompi_bitmap.c	Next step in the project split, mainly source code re-arranging	2006-02-12 01:33:29 +00:00
ompi_bitmap.h	Next step in the project split, mainly source code re-arranging	2006-02-12 01:33:29 +00:00
ompi_circular_buffer_fifo.h	* fix race condition due to poorly placed memory barriers that was causing	2006-02-16 20:14:33 +00:00
ompi_fifo.h	Next step in the project split, mainly source code re-arranging	2006-02-12 01:33:29 +00:00
ompi_free_list.c	OK my last commit was ... kind of wrong. It only worked if the element_size	2006-06-22 14:47:07 +00:00
ompi_free_list.h	If ompi_free_list_grow fails wait until resources are available instead of	2006-06-27 09:23:51 +00:00
ompi_pointer_array.c	Next step in the project split, mainly source code re-arranging	2006-02-12 01:33:29 +00:00
ompi_pointer_array.h	Update the copyright notices for IU and UTK.	2005-11-05 19:57:48 +00:00
ompi_rb_tree.c	finished the ompi_free_list_t class nightmare..	2006-06-12 22:09:03 +00:00
ompi_rb_tree.h	make ompi_free_list_item_t a class..	2006-06-12 16:44:00 +00:00
ompi_seq_tracker.c	It's not about how much memory we use, but about how we use it.	2006-07-13 04:56:15 +00:00
ompi_seq_tracker.h	- added cleanup to destructor	2006-05-04 15:14:47 +00:00