Jonathan Stone <jonathan@DSG.Stanford.EDU> writes:
> Here's a suggestion for a medium-term remedy to the lack of an kernel
> API for accessing CPU cycle-counters or high-quality time sources
> (which are inherently MD) from MI code:

One need that i've noticed implementing some things is that sometimes
you want 'microtime()' or another 'time'-based time value and other
times you want a microtime()-equivalent based on mono_time.  Some
timing users Really Don't Want time to go backwards, but have a need
for accurate time...

If we're going to add a new time API or APIs, it'd probably be at
least worthwhile to consider needs like those as well.


cgd
-- 
Chris Demetriou - cgd@netbsd.org - http://www.netbsd.org/People/Pages/cgd.html
Disclaimer: Not speaking for NetBSD, just expressing my own opinion.