lkml.org 
[lkml]   [1998]   [Dec]   [12]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
/
Date
From
SubjectRe: [NOT_A_BUG] (I think) MTRR nice things... :)
Romano Giannetti writes:
> Hi all,
>
> I was trying to x11perf(1) measure the performance boost of my ATI
> rage IIC with and without the mtrr aperture
>
> (0)pern:~/tmp% cat /proc/mtrr
> reg00: base=0x00000000 ( 0MB), size= 64MB: write-back, count=1
> reg01: base=0xaf000000 (2800MB), size= 2MB: write-combining, count=2
>
> and I did a typing error resulting in reg00 (main memory) being
> deleted. Try to guess what happened. System was _really_slow_, much
> less than my old 386sx. You could notice almost one second wait
> between prompts of shell (bash on vesafb console).

Yeah, I did this once when I was hacking the MTRR code. To my
amazement, the system didn't crash. It just was sloooooow.

> Well. I think this is a "so do not do it" thing, but I was really
> amazed to see such a _big_ slowdown. Really cache/no cache can do
> this difference?

Go ahead: do it. It's quite safe. Just not useful ;-)
The slowdown is pretty much what you'd expect. Situation normal. All
you did was to disable the cache. Now you know why we pay for caches.

For a comparison, disable the cache(s) in your BIOS setup. Note the
slowdown.

Regards,

Richard....

-
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 13:46    [W:0.042 / U:0.572 seconds]
©2003-2020 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site