Re: [openrisc] performance of openrisc

["Marko Mlinar" <markom@opencores.org>]

For code optimizations you can also use profile utility.

Marko

----- Original Message -----
From: "Damjan Lampret" <lampret@opencores.org>
To: <openrisc@opencores.org>
Sent: Wednesday, October 24, 2001 2:26 PM
Subject: Re: [openrisc] performance of openrisc


> Hi Jamil,
>
> I think the easiest way is to write the algorithm in C, compile it for
or1k
> and run it on or1ksim. This will give you approx. number of clock cycles.
>
> regards,
> Damjan
>
> ----- Original Message -----
> From: "Jamil Khatib" <khatib@ieee.org>
> To: <openrisc@opencores.org>
> Cc: <khatib@opencores.org>
> Sent: Monday, October 22, 2001 11:09 PM
> Subject: [openrisc] performance of openrisc
>
>
> > Hi,
> > I am trying to split my bluetooth project between hardware and software,
> > I have some algorithms that I think they can be implemented in software
> > but I do not have any idea about the performance of openrisc.
> >
> > The algorithms is mainly scrambler but it will be performed on several
> > steps with different polynomials. The input ot the algorithm is 24 bit
> > and the output will be 64-bit.
> > The good thing that this algorithm will not be executed frequently at
> > the worst case a new value should be generated each 312.5us.
> >
> > How can I evaluate such algorithm on openrisc for performance and power
> > consumption. This algorithm is not used all the time so it will waste
> > large amount of useless hardware.
> >
> > Thanks for your help
> > --
> >    Jamil Khatib
> > OpenCores Organization
> > http://www.opencores.org
> >
> >
> >
> > --
> > To unsubscribe from openrisc mailing list please visit
> http://www.opencores.org/mailinglists.shtml
> >
>
> --
> To unsubscribe from openrisc mailing list please visit
http://www.opencores.org/mailinglists.shtml
>

--
To unsubscribe from openrisc mailing list please visit http://www.opencores.org/mailinglists.shtml