lkml.org 
[lkml]   [1997]   [Jul]   [22]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
/
From
SubjectRe: question about M2 "feature"
Date
Ingo Molnar writes ...
>
> we could use that area for free memory handling.
>
> we could put the single-page usage bitmap there, 8k is enough to handle up
> to about 256M RAM if 1 bit represents 1 page. (a bit less than 256M RAM,
> since some bits are needed for higher order coverage bits)
>
> gosh, ~10 cycles __guaranteed__ get_free_page() execution time, ~20 cycles

I'd rather see support for letting it get used from user space. I
know other people's needs are different, but for many people, if the
kernel is using a significant amount of CPU, something is _wrong_, so
it is a waste of time making it run faster.

On the other hand, some of my computational jobs have some small regions of
memory which are _very_ hot ...

---------------------------------------------------------------------------
Tim Hollebeek | Disclaimer :=> Everything above is a true statement,
Electron Psychologist | for sufficiently false values of true.
Princeton University | email: tim@wfn-shop.princeton.edu
----------------------| http://wfn-shop.princeton.edu/~tim (NEW! IMPROVED!)

spamming idiots please email spam@franck.princeton.edu so I know who you are.
Thanks for being stupid and doing this automatically.

\
 
 \ /
  Last update: 2005-03-22 13:39    [W:0.092 / U:0.336 seconds]
©2003-2020 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site