dettmer@jupiter.informatik.uni-dortmund.de (Thomas Dettmer) (02/25/91)
Hello, Does anybody out there know, if there are plans to make an ANSI or C++ interface for CLSI's VTIP (SPI)? I've heard some rumors, but nothing seems to be sure. Background: We are using InterViews (a X11/C++ based graphic library) in some projects but we are not planning to build an analyzer. Therefore it would be very useful to use the CLSI software, but it's a lot of work to produce a C++ interface (headerfiles) for SPI. Another question related to this one: what will be the base to define a standard of the intermediate VHDL representation - a concrete language as C or ANSI C or something more abstract? And another question: As far as I can see from systems available to us, it seems not to be possible to build a set of basic libraries and simply use them in several designs/places. To make it more clearly: The lib/data managers of the systems need one directory where all the data of a design must be stored, including basic libs as STD or basic cells. On the other hand, I'd prefer separated environments/directories for each design project. Disadvantages are: 1. The whole basic stuff has to be copied (and reanalyzed). 2. After copying the basic libs, they are stuff of the designer - updates to my base have no effect and they are no longer write protected. (What I would like to have is something like -I and -L in C compilers.) Is something like this planned - or is it impossible, because the data management needs pointer like "used by"? Or am I wrong and there are systems which do act in this way? Thanks for answers tom. dettmer@jupiter.ls1.informatik.uni-dortmund.de phone: +49-231 755 4825, FAX: +49-231 755 2386 Thomas Dettmer, Dortmund University, Computer Science I Post Box 50 05 00, W-4600 Dortmund 50, Germany
marshall@manse.cs.man.ac.uk (Robert Marshall) (03/12/91)
In article <3040@laura.UUCP>, dettmer@jupiter.informatik.uni-dortmund.de (Thomas Dettmer) writes: > Does anybody out there know, if there are plans to make an ANSI or C++ > interface for CLSI's VTIP (SPI)? I've heard some rumors, but nothing seems > to be sure. > Background: We are using InterViews (a X11/C++ based graphic library) in > some projects but we are not planning to build an analyzer. Therefore it > would be very useful to use the CLSI software, but it's a lot of work to > produce a C++ interface (headerfiles) for SPI. > > Another question related to this one: what will be the base to define a > standard of the intermediate VHDL representation - a concrete language as C > or ANSI C or something more abstract? > .. I hoped there would be some response to this, but having seen none I feel obliged to say something(!) I spent some time partially converting the VIFASG 1076 VHDL Version 1.00 schema definition into C++ classes with the intention of hiding the constructor definitions into various analyser specific include files. It was great for discovering bugs in C++ compilers, at the moment I'm trying to link these classes into the CLSI system with some remaining problems (mostly connected with Sun shared libraries and g++) On the actual standard interface all that anyone seems ready to commit themselves to is a procedural interface, I don't know whether the schema defined in the document is a dead duck, what was said at Oakland in October? Various things were promised at EuroVHDL in September but I heard no more. Robert If anything was said on this (in comp.lang.vhdl) that the uk missed I'd be grateful if you could send a resume -- ------------------------------------------------------------------ Robert A.J.Marshall, EMAIL: rmarshall@cs.man.ac.uk Room 3.08, IT Building, Department of Computer Science, University of Manchester, Oxford Road, Manchester, M13 9PL, U.K. Tel: (+44) 61-275 6269 Fax: (+44) 61-275 6280 ------------------------------------------------------------------