Re: Does the KA630 have a TB?

[Mouse <mouse%Rodents-Montreal.ORG@localhost>]

> IIRC the 1.2 distribution took 11 days to compile, and 2 additional
> days to gzip the distribution files.  But it worked :-)

:-)  On what hardware, do you happen to recall?

I did a build of the (mutant 1.4T) world on my emulated VAX.  It took
about two weeks, start-to-finish, but there were some four or five
points where it failed and I had to fix something (I haven't test-built
on VAX very recently before that), so I estimate it would be about 10
days of straight running.  I intend to do a new build-of-the-world
sometime soon, so I may have something to report on that score in a
couple of weeks.  I'd _like_ to get the TB working first; it should be
_some_ performance improvement even without the JITter.

> Since (at least) 8 TBIS are needed for a kernel mapping, and even
> more for userspace, I realized it just wasn't worth it.

Why more for userspace?  If you have to change a kernel PTE, sure, but
I would expect most userspace changes would involve only P0 or P1 PTEs.
Am I wrong in thinking that?  (Admittedly, I haven't grokked the VAX
pmap in detail.)

/~\ The ASCII				  Mouse
\ / Ribbon Campaign
 X  Against HTML		mouse%rodents-montreal.org@localhost
/ \ Email!	     7D C8 61 52 5D E7 2D 39  4E F1 31 3E E8 B3 27 4B