Re: Draft 1.19 of SCE-MI pipes section of SCE-MI 2 working document

From: Per Bojsen <per.bojsen_at_.....>
Date: Tue Oct 24 2006 - 05:44:49 PDT
Hi John,

> The interface shown makes in clear in no uncertain
> terms what is required by the implementor. However
> the implementor choses to provide this interface
> is their business. Is this not true for the entire standard ?

While I agree with you, I think the issue is that if a
vendor chooses to implement this aspect of the standard using
a different mechanism than SystemVerilog interfaces (literally),
customers may complain that the vendor didn't implement the
standard and the vendor will have the burden of convincing
the users that they are indeed compliant with the standard.
So a comment or note in this area of the standard could help
reduce those support calls.

The reason this area is a special case is that it is possible
for the implementation to look significantly different from
the spec in terms of the SystemVerilog source the user might
see.

Per
Received on Tue Oct 24 05:48:32 2006

This archive was generated by hypermail 2.1.8 : Tue Oct 24 2006 - 05:48:52 PDT