On Feb 4, 2009, at 6:05 PM, Eugene Loh wrote:
- Remove a function call from the critical performance path;
possibly save a little latency
The only "benefit" is "possibly a little"? This is not at all
compelling. Is the hoped-for benefit measurable? I assume a
pingpong latency test ove
I've updated the wiki with the specific meeting location -- we're in a
different building than the MPI Forum meeting (right across the
street). For those of you who were there, it's the same conference
room that we used for the RTE meeting in December:
https://svn.open-mpi.org/trac/omp
On Sat, 7 Feb 2009, Jeff Squyres wrote:
End result: I guess I'm a little surprised that the difference is that clear
-- does a function call really take 10ns? I'm also surprised that the
layered C version has significantly more jitter than the non-layered version;
I can't really explain that.
On Feb 7, 2009, at 12:23 PM, Brian W. Barrett wrote:
End result: I guess I'm a little surprised that the difference is
that clear -- does a function call really take 10ns? I'm also
surprised that the layered C version has significantly more jitter
than the non-layered version; I can't real
