Re: My opinion re LISP

From	anton@mips.complang.tuwien.ac.at (Anton Ertl)
Newsgroups	comp.lang.forth
Subject	Re: My opinion re LISP
Date	2013-04-24 11:59 +0000
Organization	Institut fuer Computersprachen, Technische Universitaet Wien
Message-ID	<2013Apr24.135959@mips.complang.tuwien.ac.at> (permalink)
References	(1 earlier) <471240d3-0b19-4d8c-9f98-0afd6fc5c8a0@pd6g2000pbc.googlegroups.com> <7fa22bd0-de44-4b0c-ac6d-0209cf8019df@di5g2000pbc.googlegroups.com> <kl3s3v$1j5$2@online.de> <e9fa709e-e257-4d84-b077-8f3692eba44d@k6g2000pbq.googlegroups.com> <kl6t0d$1ml$1@online.de>

Show all headers | View raw

Bernd Paysan <bernd.paysan@gmx.de> writes:
>Or we do what we Forthers are used to: Try it!  The problem here is that you 
>are easily losing the forrest in the trees, as microbenchmarks may show you 
>that 16-byte-aligning is good, but for the overall program, this is indeed 
>rather bad.

Is it?  Have you measured it?

In our work on dynamic superinstructions we found surprising results,
where "optimizations" that reduced code size resulted in slower
execution through more I-cache misses.  My explanation was that the
non-"optimized" code had better spatial locality (the binary code had
the same layout as the source code).

Concerning padding after unconditional branches (after BRANCH or ;S in
Gforth), we have code for that in Gforth.  On Alphas it provides a
significant speedup, on IA-32/AMD64 I performed some measurements, and
there was little impact.  From
<http://www.complang.tuwien.ac.at/viewcvs/cgi-bin/viewcvs.cgi/gforth/arch/amd64/machine.h?rev=1.26&view=markup>:

#define MAX_PADDING 4
/* results for various maxpaddings:
   3GHz Xeon 5160                     2.2GHz Athlon 64 X2
   sieve bubble matrix  fib  padding sieve bubble matrix  fib 
    0.132 0.216  0.072 0.228    0     0.260 0.300  0.108 0.344
    0.132 0.216  0.072 0.228    1     0.268 0.300  0.112 0.344
    0.132 0.216  0.072 0.248    2     0.256 0.300  0.108 0.344
    0.136 0.216  0.072 0.248    3     0.252 0.300  0.108 0.344
    0.132 0.220  0.072 0.240    4     0.252 0.300  0.112 0.340
    0.136 0.216  0.072 0.248    5     0.252 0.300  0.108 0.344
    0.132 0.216  0.072 0.244    6     0.256 0.300  0.108 0.344
    0.132 0.216  0.072 0.244    7     0.264 0.300  0.108 0.344
    0.136 0.216  0.072 0.244    8     0.268 0.296  0.108 0.340
*/

- anton
-- 
M. Anton Ertl  http://www.complang.tuwien.ac.at/anton/home.html
comp.lang.forth FAQs: http://www.complang.tuwien.ac.at/forth/faq/toc.html
     New standard: http://www.forth200x.org/forth200x.html
   EuroForth 2013: http://www.euroforth.org/ef13/

Back to comp.lang.forth | Previous | Next — Previous in thread | Next in thread | Find similar

Thread

re: My opinion re LISP "WJ" <w_a_x_man@yahoo.com> - 2013-04-16 15:50 +0000
  Re: My opinion re LISP Hugh Aguilar <hughaguilar96@yahoo.com> - 2013-04-20 21:20 -0700
    Re: My opinion re LISP Hugh Aguilar <hughaguilar96@yahoo.com> - 2013-04-21 16:46 -0700
      Re: My opinion re LISP Bernd Paysan <bernd.paysan@gmx.de> - 2013-04-22 19:30 +0200
        Re: My opinion re LISP Hugh Aguilar <hughaguilar96@yahoo.com> - 2013-04-22 21:41 -0700
          Re: My opinion re LISP Bernd Paysan <bernd.paysan@gmx.de> - 2013-04-23 23:04 +0200
            Re: My opinion re LISP Ron Aaron <rambamist@gmail.com> - 2013-04-24 06:10 +0300
            Re: My opinion re LISP Hugh Aguilar <hughaguilar96@yahoo.com> - 2013-04-23 22:17 -0700
            Re: My opinion re LISP anton@mips.complang.tuwien.ac.at (Anton Ertl) - 2013-04-24 11:59 +0000
      Re: My opinion re LISP Hugh Aguilar <hughaguilar96@yahoo.com> - 2013-04-22 21:49 -0700

csiph-web