From: Adam Krolnik (krolnik@lsil.com)
Date: Tue Jun 18 2002 - 13:37:43 PDT
Precedence: bulk
Hi Steve;
I spoke to the writer of the comments.
>On a side note, does it seems that several assumptions
>need to be made to make the library declarations
>workable in a real environment:
> 1. paths ending with /*.v (wildcard anywhere in
> the base of the filename) contain only the
> module with the name matching the filename base.
His initial thoughts were based on using the library map files
to find a module (more efficiently.) I talked with him about one
model where the map file is used to define the set of modules belonging
to
a library (which then requires some amount of file parsing.)
This is one common item in design guides, though it is not necessary.
It makes it easier for people definitely.
2. paths with a wildcard in the extension
must match a "default verilog" filename extension
and only contain the module with the name matching
the filename.
The writer uses a design methodology where multiple types of source code
exists
in one directory. Some methodologies do this, some separate source code
by type.
Neither is excluded by the standard.
His comment about block comments is correct though - what happens?
We will have to discuss the options when we review the file_path_spec
erratta.
Adam Krolnik
Verification Mgr.
LSI Logic Corp.
Plano TX. 75074
This archive was generated by hypermail 2.1.4
: Mon Jul 08 2002 - 12:55:37 PDT
and
sponsored by Boyd Technology, Inc.