Re: [microblaze-uclinux] OT?, Performance

[Matthew Rubenstein <email@xxxxxxxxxxxx>]

On Thu, 2004-05-13 at 18:03, John Williams wrote:
> emanuel stiebler wrote:
> > Goran Bilski wrote:
> > 
> >> Another way to improve the numbers is to use the harvard architecute 
> >> of MicroBlaze and have 2 OPB busses and two external memories.
> > 
> > 
> > But we can't support it on the uclinux ...
> 
> That's not strictly true...  it would just require some (major?) work in 
> the kernel, to support two memory pools - one for data, and one for 
> code, and in the hardware we'd instantiate bridges so that we are able 
> to write into code space.
> 
> Not trivial, but certainly doable I think.

	Another benefit of the MB/uCL platform over pure sequential logic is
revising the VM to secure memory access. Implementing allocated buffers
in HW could eliminate "buffer overflow attacks", while vectoring them to
an interrupt could allow the OS to trap them as exceptions without
development or runtime performance overhead penalties. Similarly for
code segmentation faults. And such an architectural improvement could
offer a platform for securing processes against unauthorized access to
one another's code or data, again in the OS without performance
penalties.


> John
> 
> 
> ___________________________
> microblaze-uclinux mailing list
> microblaze-uclinux@itee.uq.edu.au
> Project Home Page : http://www.itee.uq.edu.au/~jwilliams/mblaze-uclinux
> Mailing List Archive : http://www.itee.uq.edu.au/~listarch/microblaze-uclinux/
-- 


(C) Matthew Rubenstein

___________________________
microblaze-uclinux mailing list
microblaze-uclinux@itee.uq.edu.au
Project Home Page : http://www.itee.uq.edu.au/~jwilliams/mblaze-uclinux
Mailing List Archive : http://www.itee.uq.edu.au/~listarch/microblaze-uclinux/