On Wed, Mar 27, 2002 at 03:33:03PM -0500, Marcus Brinkmann wrote:
> On Thu, Mar 28, 2002 at 10:12:12AM -0700, Jon Arney wrote:
> > Here's a start at some performance statistics. I suggest that
> > if anyone else is interested in characterizing the system that
> > they begin with downloading bonnie:
> >
> > http://www.textuality.com/bonnie/download.html
> >
> > Run it under the Hurd and Linux (on the same box) and send the results
> > to me.
>
> Last thing I heard is that bonnie crashes the Hurd, but hey, you can fix
> the bugs and _the_ send the results to Jon :)
>
> At some time, I made sure that profiling works in the Hurd. So far,
> nobody has seriously profiled the Hurd servers, libraries and glibc.
> We really would like to see people profiling things and telling us where
> the bottlenecks are.
I've already done this, and I post the results on this list. Check it out...
BTW it seems that profiling servers doesn't give coherent results: they give
different results just making apparently insignificant modification to the
code. Watching the code of glibc & gnumach, it seems that system call used
by gprof'ed executable take in account only a thread so results can be not
completly significative. A mono thread (or bi-thread) server can help in
profile it, it would better than nothing.
PS: I've also already post some results and comparison of hurd vs linux
using bonnie++.
--
Saluti / Regards
Diego Roversi |
| diegor at tiscalinet.it
_______________________________________________
Bug-hurd mailing list
[EMAIL PROTECTED]
http://mail.gnu.org/mailman/listinfo/bug-hurd
