cvs commit: src/sys/i386/i386 pmap.c

[Robert Watson]

On Tue, 9 Nov 2004, Julian Elischer wrote:

> Is this with HTT? 
> 
> We've dicoverred that HTT makes better code run slower, remember? 

Nope, the issue here is a minor tweak in the spin lock code that causes
moving away from an atomic operation to make it run slower rather than
faster.  I think John has a theory that the lack of using an atomic
operation slows the rate at which the change is visible on other
processors by virtue of being non-atomic, so locks contend longer.

Robert N M Watson             FreeBSD Core Team, TrustedBSD Projects
robert at fledge.watson.org      Principal Research Scientist, McAfee Research