Hi;
[EMAIL PROTECTED] wrote:
> hm, I use write8 to fill up memory, still very slow (my CPU is AMD K6-350)
> -_-.
> I prefer to write some routines to fill up using native implementation.
Two suggestions:
(1) Find out why the write* routines are so slow (I can think of several reasons
right off of the top of my head... why not just re-order the built-in-routine
checks to check the read/write operations first, before the other ones?), and
(2) See if the JavaOS "memory" classes have any interfaces for block memory
operations, and then implement them.
> >I have the same problem: running a long task is impossible. (I think it
> >was something with jbHeap())
> >
>
> I think so. I think there is one more reason. We use lots of write8. But we
> havn't
> delcare a variable to get the value from write8. It may cause some problems.
I think we run out of memory after awhile, due to "leaking" Java stack frames.
This is why George's new conservatively-GCed heap implementation is so
time-critical.
-jm (still in crisis mode)
--
==== John Morrison ==== MaK Technologies, Inc.
==== Chief Technology Officer ==== 185 Alewife Brook Pkwy, Cambridge, MA 02138
==== [EMAIL PROTECTED] ==== http://www.mak.com/welcome.html
==== vox:617-876-8085 x115 ==== fax:617-876-9208
_______________________________________________
Kernel maillist - [EMAIL PROTECTED]
http://jos.org/mailman/listinfo/kernel
