[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: [f-cpu] New snapshot for EU_INC and EU_CMP
Hello,
Etienne LABARRE wrote:
>Hi all,
>
<cut>
>The precise latency is not important,
> because max latency will be fixed by timing constraints during synthetisis process.
> (It's an explain of my choice of standard function. "Just an
> Illusion"
> will can explain this more easy that me, i think...)
>
In many synthesis tools you can define some critical path and give
timing constraint to the optimization process, in the current case we
couldn't predict the latency time in general because that depend of the
optimization result , and the target technology.
We can ensure a pseudo latency timing if we manually force the mapping
(use specific 4-and components by example), but in that case our code
couldn't really migrate between different target technology.
And this type of constraint (if they are generalized) can work against
optimization process.
The avantage of using normalized, or by default standardized, functions
is than synthesizer have generally yet optimized this functions and can
ensure best result with any target technology.
>Etienne.
>
Cheers
Just an Illusion
--
______________________________
"The matrix is my world, I am a shadow.
Shadow in world, shadow in life. Don't try to keep me,
I am a Corpo's Killer.
Don't follow me or die..."
The KingWalker - 1996
*************************************************************
To unsubscribe, send an e-mail to majordomo@seul.org with
unsubscribe f-cpu in the body. http://f-cpu.seul.org/