Subject: Re: Netscape.
To: None <current-users@NetBSD.ORG>
From: Chris G Demetriou <Chris_G_Demetriou@LAGAVULIN.PDL.CS.CMU.EDU>
Date: 03/12/1995 00:58:54
> I'm not sure what the issues are surrounding converting the
> hp300 port (the port that I use most) to 8k pagesize
> so it can run binaries from the other 68k ports...
Then Charles said:
> It's not necessary to `convert' the port in order to do that.
to elucidate a bit:
To get the hp300 (the only m68k4k port) to properly work with m68k8k
binaries, one needs only to add a hook for a machine-dependent exec
type, which knows how to do the binaries' mappings correctly. That
hook is relatively easy to write.
It's a bit harder to write a similar hook for m68k8k ports, to allow
them to run the hp300 (m68k4k) binaries. In fact, it's impossible
to write a "m68k8k machine runs m68k4k binaries" hook to run standard,
demand-paged ZMAGIC binaries that is as memory- and i/o-efficient as
is running them on the hp300.
some (including me 8-) would argue that the hp300 should be converted
to use 8k page sizes for all of its binaries. This would allow hp300
and other m68k ports' binaries to be completely interchangeable, with
no special hooks, etc. It would cost (on average) 4k extra disk space
per binary, on hp300 systems. That cost comes out to 2-3M, given an
average extra space per binary of 4k, and given between 500 and 750
binaries per system (the average, i'd say, at least looking at one of
the hp300's i've access to).
I think that that waste of space is worthwhile, because it allows:
(1) all m68k ports to use binaries interchangably, with no special
hooks, options, or ineffiencies,
(2) hp300's to easily (by default, even!) compile binaries
that will run on the other m68k ports, and vice-versa.
Last i checked, this wasn't possible, and was bloody