[Author Prev][Author Next][Thread Prev][Thread Next][Author Index][Thread Index]

Re: [f-cpu] F-CPU SoC



Le 2015-04-01 23:01, Nikolay Dimitrov a ÃcritÂ:
Hi Yann,
good morning,

Sorry. I think I used the term properly, it's just that I didn't
explained my thoughts. What I meant was that there's no one "true F-CPU
SoC". The SoC will use a profile-specific implementation of the F-CPU
ISA. And yes indeed, I'm talking about FPGA.

Speaking of profile :
I've implemented a system to manage that with the YASEP.
It's very handy ! It must be more stable for F-CPU but
the "profiles" are not just an idea and few other CPU
can boast this.

I have some extra-pervert ideas for the tooling, but need some
more time before giving actual proposal :D.
please elaborate :-)
I'll do when I order my thoughts on the subject.
I can't wait !

I have several "starter kit" boards, one with the same AFS600 indeed,
so I could prototype a few stuff, but also more powerful boards with
AFS1500, which might be ok for early F-CPU implementations. However
the I/Os are minimal, there is no external RAM, the interfaces are
lacking... so it can't implement a whole system.
These FPGA are probably fine for the ISA development, but for working
on actual SoC the lack of memory interfaces (or enough I/Os to do it
the hard way), memory will be a limiting factor, imho.

The AFS1500 is a little beast so I bought several
boards for workshops to present the YASEP.
But there is room for a reasonable F-CPU
(though the internal wire delays won't allow it to run fast)
There are enough I/O pins so an external
memory board could be hacked (I have a bunch of SRAMs).

The internal Flash is 1MB and can feed a RISC CPU's instruction decoder at 100MHz
http://www.microsemi.com/products/fpga-soc/fpga/fusion#product-tables
There's enough SRAM for data cache.

So it's better than nothing for F-CPU development.

The chance is that the majority of people will use whatever low-cost
hobby board is available on the market, so there's zero probability that
these boards will be compatible with such common interface.
For F-CPU, there is a need of a fast 64-bits wide data bus.
Few hobbyist boards provide this.

In practice, all hobby FPGA boards will have somewhat different power
supplies, clocks, programming interfaces, pinouts/connectors. So I would
just leave it as is - let the guy who is porting the F-CPU SoC to this
specific board to take care of the board specifics and to document which
SoC signal goes where, and that should work OK.

yup, however for our team's progress, we need a uniform platform
because otherwise we'd lose a lot of time dealing with each person's
board details...

So my proposal for F-GPU is to prototype it then
build 8 or 12 of them for the team's most active developers.
The cost would be bearable and the risk is low
because the F-GPU is repurposable.

If there's a need for interaction between several (possibly different)
FPGA boards, it would be easy to hack a homebrew cables, as I don't see
our designs soon to be clocked so high that we couldn't make cables for
them at home :D.

"assume nothing" ! :-D
I got tricky issues even when I didn't expect them...

Regards,
Nikolay
YG
*************************************************************
To unsubscribe, send an e-mail to majordomo@xxxxxxxx with
unsubscribe f-cpu       in the body. http://f-cpu.seul.org/