Re: mmap() is slower than read() on SCSI/IDE on 2.0 and 2.1

Marc Lehmann (pcg@goof.com)
Sat, 19 Dec 1998 21:37:19 +0100

On Sat, Dec 19, 1998 at 04:37:34PM +0100, Rogier Wolff wrote:
> Marc Lehmann wrote:
> > On Thu, Dec 17, 1998 at 06:52:50AM +0000, Linus Torvalds wrote:
> > > Umm, the easiest hint is probably to just look at the faulting address.
> > > We have it available, after all.
> > >
> > > I suspect that such a simple heuristic would be fairly accurate, and it
> > > can be coupled with other heuristics to further increase the accuracy.
> >
> > file copy, yes. But grep (and probably lots of others) don't access memory
> > sequentially (as faster search algorithms exist)
>
> A fast search algorithm, touches memory every n bytes where n is the
> size of the largest constant string that you're searching for.

A a small threshold might catch fgrep, but it will also increase the
possibility for false "hits".

Other (important) types of searches (for example a linear search in a
database table) might not offer such behaviour.

In any case, that makes an excellent heuristic if combined with watching page
access ordering.

And it might well give a good heuristic when used alone (but I'd benchmark it
;)

--
Happy New Year, I'll be away from 21. Dec to 7. Jan


      -----==-                                              |
      ----==-- _                                            |
      ---==---(_)__  __ ____  __       Marc Lehmann       +--
      --==---/ / _ \/ // /\ \/ /       pcg@goof.com       |e|
      -=====/_/_//_/\_,_/ /_/\_\                          --+
    The choice of a GNU generation                        |
                                                          |


-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.rutgers.edu
Please read the FAQ at http://www.tux.org/lkml/