9fans - fans of the OS Plan 9 from Bell Labs
 help / color / mirror / Atom feed
* Senseless benchmarks
@ 1994-03-28  7:52 David
  0 siblings, 0 replies; 4+ messages in thread
From: David @ 1994-03-28  7:52 UTC (permalink / raw)


> > I just did that on an SS-1000 running SunOS 5.2, and it took 17.3 seconds.
> 
> Make sure you statically link it.  Otherwise you pay for all sorts
> of shared library overhead.

I just tried it, and it runs twice as fast (around 8 seconds) if I
statically link.  Plan 9 on a lowly sparc ELC runs the test in about
4 secs -- twice as fast again!

I heard some rumor about fork on Solaris taking 1/10th of a second
around the time that we got the SS-1000s, but perhaps this was
fixed in a later release of Solaris.




^ permalink raw reply	[flat|nested] 4+ messages in thread

* Senseless benchmarks
@ 1994-03-28  6:06 Scott
  0 siblings, 0 replies; 4+ messages in thread
From: Scott @ 1994-03-28  6:06 UTC (permalink / raw)



> I just did that on an SS-1000 running SunOS 5.2, and it took 17.3 seconds.

Make sure you statically link it.  Otherwise you pay for all sorts
of shared library overhead.





^ permalink raw reply	[flat|nested] 4+ messages in thread

* Senseless benchmarks
@ 1994-03-28  5:49 Piers
  0 siblings, 0 replies; 4+ messages in thread
From: Piers @ 1994-03-28  5:49 UTC (permalink / raw)


	Date:   Mon, 28 Mar 1994 00:27:03 -0500
	From:   Scott Schwartz <schwartz@groucho.cse.psu.edu>
	Message-Id: <94Mar28.002712est.2542@groucho.cse.psu.edu>

	Just for fun, I ran the following program on a machine running SunOS and on a
	machine running Plan 9 (with obvious alterations to get it to compile.)
	I was hoping to see Plan 9 go faster, but they took about the same time
	(10 seconds on a Sparcstation I).
	...

I just did that on an SS-1000 running SunOS 5.2, and it took 17.3 seconds.




^ permalink raw reply	[flat|nested] 4+ messages in thread

* Senseless benchmarks
@ 1994-03-28  5:27 Scott
  0 siblings, 0 replies; 4+ messages in thread
From: Scott @ 1994-03-28  5:27 UTC (permalink / raw)


Just for fun, I ran the following program on a machine running SunOS and on a 
machine running Plan 9 (with obvious alterations to get it to compile.)
I was hoping to see Plan 9 go faster, but they took about the same time
(10 seconds on a Sparcstation I).  Has anyone done more thoughtful
benchmarks comparing the two kinds of systems?
  
/*
#include <u.h>
#include <libc.h>
#define exit exits
*/

main() 
{
  int i;
  for (i=0; i<1000; ++i) {
    if (fork()) {
      wait(0);
    } else {
      exit(0);
    }
  }
}

P.S.  reason #2 why plan 9 is not my favorite operating system "vc, xc, etc".
given that you will usually use an mkfile most of the time anyway, 
why not name them something sensible like "sparc-cc, mips-cc, etc-cc"? 







^ permalink raw reply	[flat|nested] 4+ messages in thread

end of thread, other threads:[~1994-03-28  7:52 UTC | newest]

Thread overview: 4+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
1994-03-28  7:52 Senseless benchmarks David
  -- strict thread matches above, loose matches on Subject: below --
1994-03-28  6:06 Scott
1994-03-28  5:49 Piers
1994-03-28  5:27 Scott

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).