> 
> >>>>> "Jon" == Jon Buller <jonb@metronet.com> writes:
>     Jon> That sounds real slow, does EVERY SPARC kernel do that on ALL
>     Jon> traps?  It sounds like a config option that would inline the
>     Jon> code for MY machine and eliminate the code for all others
>     Jon> would be significantly faster.  (All this from someone who
>     Jon> has not looked at the code, but did get an ELC a few months
>     Jon> ago and is currently running said code.)
> 
>   Since these traps are pointed to by a trap table, wouldn't it make
> the most sense to just point at different trap handlers at run-time?

That's exactly what happens right now. The reason that the hypersparc
complicates issues is that you would then need a separate trap table
for each supersparc and hypersparc rev. that handled the AFSR differently.
Also, I think the microsparc handles it differently as well...

It just seems gross to build a second trap table for each CPU rev, for
the sake of a single trap. But we certainly use different tables for 
Sun4/4c/4m.

--Aaron