[En-Nut-Discussion] network throughput gain (nicrtl output)

[Ralph Mason]

I am planing on converting to a Oki ARM processor sometime early in 2004, I
expect very much that I will stick with NutOS for that transition.

I haven't looked into the tool chain that much, other than to say it will be
GCC in some form or another.

The case for a Mega128 for running NutOS is not so good as it was a few
years ago.  You can get many more mips for the same money from an ARM, and
there are also lots to choose from.


Ralph


> -----Original Message-----
> From: en-nut-discussion-bounces at egnite.de
> [mailto:en-nut-discussion-bounces at egnite.de]On Behalf Of Damian Slee
> Sent: Monday, 1 December 2003 1:42
> To: Ethernut User Chat (English)
> Subject: RE: [En-Nut-Discussion] network throughput gain (nicrtl output)
>
>
> Hi,
> anyone know of a good ARM9/ARM7 GCC toolchain?
> preferably compiled to use under windows, Like WinAVR.  If there
> was one around, more of us probably would use an ARM.
>
> There are some nice ARM processors out there with lots of flash
> and RAM on chip.
>
> -----Original Message-----
> From: Ralph Mason [mailto:ralph.mason at telogis.com]
> Sent: Monday, 1 December 2003 8:02 AM
> To: enut at ixo.de; Ethernut User Chat (English)
> Subject: RE: [En-Nut-Discussion] network throughput gain (nicrtl output)
>
>
>
> If your using a CPLD why not just do a DMA mode?
>
> Set a Read address and a write address, and a number of bytes and set it
> going, with an output done line.
>
> You could clock it from the same crystal as the AVR and do a byte
> per clock.
> Then in the AVR code you set it up and wait for the done line.
>
> But if you do all this why not just use a faster chip some ARM variant
> perhaps, and get rid of the CPLD etc altogether.  They are as cheap as an
> mega128 these days anyway.
>
> Ralph
>
>
> > -----Original Message-----
> > From: en-nut-discussion-bounces at egnite.de
> > [mailto:en-nut-discussion-bounces at egnite.de]On Behalf Of Waschk,Kolja
> > Sent: Monday, 1 December 2003 10:55
> > To: Ethernut User Chat (English)
> > Subject: Re: [En-Nut-Discussion] network throughput gain (nicrtl output)
> >
> >
> > > If I'm not wrong, this is even 1 cycle faster.
> >
> > > .L94:
> > >  ld r24,Z+
> > >  sts -16376,r24
> > >  subi r22,lo8(-(-1))
> > >  sbci r23,hi8(-(-1))
> > >  brne .L94
> >
> > 4+4+1+1+2 = 12 cycles per byte, whereas
> >
> > >> .L94:
> > >>   ld r24,X+
> > >>   std Y+16,r24
> > >>   subi r18,1
> > >>   brcc .L94
> > >>  subi r19,1
> > >>  brcc .L94
> >
> > 4+4+1+2 = 11 cycles most of the time,
> > 4+4+1+1+1+2 = 13 cycles only at 8 bit boundary
> >
> > For 536 bytes that's about 9% gain
> >
> > To push throughput even more, I'm currently playing with the idea of
> > implementing an operating mode where m128 issues a read from RAM
> > and some logic
> > (CPLD) translates this into a write access to the MAC at the same
> > time (RAM
> > outputs data, MAC reads it).  Thus no explicit write
> instruction would be
> > needed, and the inner loop would shrink to 7 cycles. Anyone tried
> > something
> > similar before?
> >
> > Kolja
> >
> > --
> > mr. kolja waschk - haubach-39 - 22765 hh - ger
> > phone +49 40 889130-34 - fax -35 - e-mail s.a.
> >
> > _______________________________________________
> > En-Nut-Discussion mailing list
> > En-Nut-Discussion at egnite.de
> > http://www.egnite.de/mailman/listinfo.cgi/en-nut-discussion
> >
>
> _______________________________________________
> En-Nut-Discussion mailing list
> En-Nut-Discussion at egnite.de
> http://www.egnite.de/mailman/listinfo.cgi/en-nut-discussion
> _______________________________________________
> En-Nut-Discussion mailing list
> En-Nut-Discussion at egnite.de
> http://www.egnite.de/mailman/listinfo.cgi/en-nut-discussion
>