lkml.org 
[lkml]   [1998]   [Nov]   [11]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
/
Date
From
SubjectRe: TSC and Cyrix
Kurt Garloff wrote:
>
> On Sun, Nov 08, 1998 at 02:42:03PM -0600, Trever Adams wrote:
> > Supposedly, in 2.1.127 it should output Mhz if the CPU supports TSC
> > (note that bogomips is pretty accurate to the Mhz with Cyrix chips, but
> > that is not what I am after). I have seen a tremendous speed up in all
> > apps (multitasking not just the foreground process) with 2.1.127, so I
> > assumed that the TSC code was active... now I found out it doesn't seem
> > to be.
>
> Why do you expect the TSC code to speed up your apps? It just makes the
> timing more accurate, if I didn't get anything completely wrong.
>

This is what I have understood from following the TSC additions etc:
Every time a time type function is called from userspace (or in the
kernel) it has to either use the TSC, or go CPU to the 85xx (I think
that is the number anyway) and fetch the time. It was my understanding
that this was some ridiculously long period (for computers). I also
seem to remember David M. saying that the networking code does work a
bit faster with TSC. My non-perfect tests also would support that...
but that could also be caching issues since I do not have two identical
boards with one TSC usable chip (pre 2.1.127) and one not.

Again, I may be flawed, but that is what I understood.

Trever

-
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/

\
 
 \ /
  Last update: 2005-03-22 13:45    [W:0.068 / U:0.576 seconds]
©2003-2020 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site