[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Symbolics vs stock hardware
Date: Fri 17 Feb 89 03:01:38-PST
From: TYSON@Warbucks.AI.SRI.COM (Mabry Tyson)
I think Symbolics' storage management system is considerably more
flexible than conventional systems currently have (as far as I know,
but I haven't studied this area). I doubt many of us take much
advantage of it. I don't know how much of it could be done with
Yes, it's great! But you can play similar schemes on stock hardware;
Re CDR coding. One advantage CDR coding (only reasonably available
with hardware assist) is the compaction of memory usage. This pays back
in dividends of lower paging time.
What's "reasonably available?" SPICE Lisp cdr-coded (on the VAX at least).
Re Forwarding pointers. The incremental GC that forwarding
pointers gives you is a nice idea--however, I can't ever recall really
needing it. I find that the dynamic GC on the Symbolics interferes so
much that we never use it. The other uses of forwarding pointers,
as outlined by Reti, are much more useful. I hope that stock hardware
will someday include these.
Forwarding pointers are a big win in terms of the ability to do
structure-forwarding while debugging (I also don't use the Dynamic GC).
But they're a loser if you want to depend on them heavily, since they
cause a trap which can cause your pipe to become invalid. I talked to
some hardware people out here; it seems unlikely that invisible pointers
will be common any time soon because of the hair involved (since today's
fashion is RISCy, hair of this sort is cut short).
(Did the Dec-10 have these, or was that
only for certain types of indirect addressing?)
Right. If you specify indirect addressing on the PDP-10 (please don't
use the past tense!) the target word will be fetched and then full
address calculation performed (including indexing and indirect
addressing again). This always has scared me since you can get into
infinite loops this way.