Jonathan Stone <jonathan@DSG.Stanford.EDU> writes:
> But still, I'm not handwaving: I squeezed the original pmax
> bootblocks into 7.5K, and. (arguably the original source of the
> no-symlink ifdefs) and talked at some length with Simon about the
> 2-stage pmax bootblocks.

... and the changes that made them fit on the pmax didn't even get
particularly close to making them fit on the alpha.  i don't recall
how much they were over, but it was a bunch (i _maybe_ recall that it
was on the order of at least 50%, but i could be OTL); alpha code is
significantly larger than mips code.

take a look at the commit msg on e.g. lib/libsa/ufs.c rev 1.23 to see
what it took to make them fit on the alpha.  8-)



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