On Thu, 24 Oct 1996 10:45:50 -0500 
 Miguel de Icaza <miguel@nuclecu.unam.mx> wrote:

 > Dave actually goes and measures the speed of his code before commiting
 > it to the tree.  He does not go and replaces a C routine with an
 > assembly routine that actually performs slower (sound familiar?  the
 > NetBSD checksum code had this very problem, I think Theo already
 > pointed this out to you guys).

Sound familar?  Uhh, not really...

The only converstaion I recall about in_cksum() was about a `faster'
i386 version.  After an analysis of the test results, the conclusion
was reached that the tests were faulty, and in fact the `faster' version
was actually slower than the one in the tree... Thus, the `faster'
version was not committed to the tree.

We recently got a new in_cksum() for the sparc which is a fair bit faster
than the old one.

Jason R. Thorpe                                       thorpej@nas.nasa.gov
NASA Ames Research Center                               Home: 408.866.1912
NAS: M/S 258-6                                          Work: 415.604.0935
Moffett Field, CA 94035                                Pager: 415.428.6939