In article <1992Aug17.065450.28834@colorado.edu>, drew@ophelia.cs.colorado.edu (Drew Eckhardt) writes:

| Boom.  You've just dirtied that page, and it is no longer 
| shareable.  So, instead of having > 200K shared between all
| of your shells, you have closer to 800K (Who has less than
| 4 shells running at a time).  Not a good thing (tm).

  Well, no, actually. Since the new version of the page is still valid
for all processes it would still be sharable. And it would not have to
be saved somewhere in the unlikely event that it gets paged out, because
the unmodified varsion is still perfectly valid.

  Since there is a choice of using -jump or not, I'm not suggesting
that this is needed, but I certainly think that it does not present the
technical problems you imply. The argument that the kernel does not
currently support this is valid, and I don't suggest that support needs
to be added. As long as there is a way other than static linking to get
around the use of the jump table for time critical programs, that
solves the problem, at least as I see it.

  Having used runtime linking on a real pig of a computer, I see not
reason why that would not be satisfactory in terms of performance, but I
also don't see that's it's critical at this time. If I ever feel the
need for it, I'll write it. I did it once, it's tedious but not
challenging, and adds as much overhead to short programs as jump tables
to to long programs.

-- 
bill davidsen, GE Corp. R&D Center; Box 8; Schenectady NY 12345
    I admit that when I was in school I wrote COBOL. But I didn't compile.