Re: [f-cpu] SIMD and exception

[Yann Guidon <whygee@f-cpu.org>]

hi,

gaetan@xeberon.net wrote:

> [...]
sorry i can't follow everything ...

> ok thank you very much. !!
> If i count well, i'm far from 4 stages for fadder... it will take at
> least takes 6 cycles... :'(
i don't think that it's a rally critical issue.
if it works correctly and can be deeply pipelined,
it will be very valuable for the F-CPU, unlike
other "off the shelf" IP cores.

I have read small parts of Kahn's work
(from the URL that you gave) and i have found
interesting explanations : for example the
progressive underflow managed by inserting
pseudo-instructions in the pipeline, or detecting
risks of underflow in advance with the MIPS method.
That's really interesting.

YG

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