Hello,
I am in the process of writing some VLSI CAD applications
in Lisp.
I would like to integrate them all to form a meaningful
software, for analysis of circuits and VLSI layouts.
I am looking for some pointers (code, description, doc)
for the following areas :
1. File parsing : recursive-descent would do
( I mean , if LALR generator is not there, then I
will do recursive-descent). The type of files I would
read are VHDL, Verilog and stuff.
PS: Yes I know EDIF would be best :)
2. Numerical Algorithms in Lisp (Matrix operations like EigenValue)
3. Geometric Algorithms (line intersection is one example)
some thing like a code repository. If all this is not
there, then I may have to write at least that part
which I need, and I will post it some where.
But my Lisp coding is not THAT good.....
4. Simple and fast GUI implementation for X-Windows.
Preferable NOT using Tcl/Tk.
What options are available. CLIM and CAPI
are not working on my system.
Liquid Common Lisp 5.0.3
5. I also need some info on "deploying" applications.
I have Harlequin Liquid Commom Lisp 5.0.3. -> Primary development
and hopefully deployment also.
Also I have CLISP, CMU-CL Python and GCL. Allegro 5.0.1
and LispWorks Trial Editions also.
OS are Linux, Win 98 and HP-UX 11.00 -> primary dev.
I think that CMU-CL Python is _quite_ good, and it runs well.
So I am concentrating on CMUCL for now : I always have Liquid
Common Lisp for really BIG applications.
I would like to hear from other Lisp developers here who have
experience in developing CAD applications for VLSI.
I look forward to hearing any/all comments on this NG.
Regards,
Sandeep Koranne
--
"Lets Make Things Better"
ED&T Test, WAY 3.37
Philips Research Laboratories Botstraat, 7
5656 AA, Eindhoven 5654 NL, Eindhoven
The Netherlands The Netherlands
Phone: +31-(040)-27 45250 +31-(040)-2573492
Fax : +31-(040)-27 44626
E-mail: ···············@philips.com
I have a similar desire to port a lot of existing Symbolics CAD tools to
a common platform.
[Give me a week or so and I'll start putting some code out on the www. I
just finished our
WWW site least week. (www.analogsystems.com) but there is no link to the
LISP stuff.
I'll add one soon and start putting the self-contained stuff out there]
Sandeep Koranne wrote:
> Hello,
> I am in the process of writing some VLSI CAD applications
> in Lisp.
>
> I would like to integrate them all to form a meaningful
> software, for analysis of circuits and VLSI layouts.
>
> I am looking for some pointers (code, description, doc)
> for the following areas :
>
> 1. File parsing : recursive-descent would do
> ( I mean , if LALR generator is not there, then I
> will do recursive-descent).
I don't know what an LALR is so I doubt I have one of those!
> The type of files I would
> read are VHDL, Verilog and stuff.
> PS: Yes I know EDIF would be best :)
>
> 2. Numerical Algorithms in Lisp (Matrix operations like EigenValue)
I downloaded a lot of these in the past: they exit already. Someone more
familiar with the newsgroups can probably give you the sources "fresh"
off the WWW.
If you have any trouble let me know and I'll package up and email you
mine. (Mine is by Gerald Roylance circa 1986 and he allows
non-commercial distribution).
[Depends what you want to do but I call up a circuit simulator (like
Spice) using RPC calls and have that crank all the math for an intense
"tightly coupled" analog circuit so i, in effect, use the matrix package
in Berkeley Spice.]
> 3. Geometric Algorithms (line intersection is one example)
>
> some thing like a code repository. If all this is not
> there, then I may have to write at least that part
> which I need, and I will post it some where.
> But my Lisp coding is not THAT good.....
>
> 4. Simple and fast GUI implementation for X-Windows.
> Preferable NOT using Tcl/Tk.
> What options are available. CLIM and CAPI
> are not working on my system.
> Liquid Common Lisp 5.0.3
Decided to use CLIM on Linux. Would be interested to hear about any Win
options. [I started to do a presentation system on ACL 5.0.1 copying the
approach of Genera but I am only one guy and ran out of steam...].
> 5. I also need some info on "deploying" applications.
> I have Harlequin Liquid Commom Lisp 5.0.3. -> Primary development
> and hopefully deployment also.
>
> Also I have CLISP, CMU-CL Python and GCL. Allegro 5.0.1
> and LispWorks Trial Editions also.
> OS are Linux, Win 98 and HP-UX 11.00 -> primary dev.
>
> I think that CMU-CL Python is _quite_ good, and it runs well.
> So I am concentrating on CMUCL for now : I always have Liquid
> Common Lisp for really BIG applications.
>
> I would like to hear from other Lisp developers here who have
> experience in developing CAD applications for VLSI.
w.r.t deploying applications: I have deployed an ACL application within
a prior company. It was to run verification on a piece of ATE (automatic
test equipment for the IC industry) it worked very well. used the
Allegro "make an application" code and after one or two false starts got
it going.
Note: my idea of LISP CAD always included the compiler because my style
is to extend LISP to have CAD features. [A simple example of which is a
logarithmic loop path so you can do AC kinda things in LISP code, a more
complex example of which I alluded to in an earlier NG message -
compiling a logic gate representation at run time] This runs into
problems with Allegro's license since I keep wanting to add the compiler
(and the compete IDE for that matter!) to the deliverable.
> I look forward to hearing any/all comments on this NG.
keep in touch: this is a great interest of mine.
> Regards,
> Sandeep Koranne
>
> --
> "Lets Make Things Better"
> ED&T Test, WAY 3.37
> Philips Research Laboratories Botstraat, 7
> 5656 AA, Eindhoven 5654 NL, Eindhoven
> The Netherlands The Netherlands
> Phone: +31-(040)-27 45250 +31-(040)-2573492
> Fax : +31-(040)-27 44626
> E-mail: ···············@philips.com
On Mon, 6 Mar 2000 10:08:29 GMT, Sandeep Koranne
<·······@natlab.research.philips.com> wrote:
>some thing like a code repository. If all this is not
>there, then I may have to write at least that part
>which I need, and I will post it some where.
>But my Lisp coding is not THAT good.....
Why not starting a code repository instead? O:-)
>4. Simple and fast GUI implementation for X-Windows.
>Preferable NOT using Tcl/Tk.
>What options are available. CLIM and CAPI
>are not working on my system.
>Liquid Common Lisp 5.0.3
Amulet? I think that's the name of garnet's sucessor, but I'm
not sure...
Good luck! :-)
//-----------------------------------------------
// Fernando Rodriguez Romero
//
// frr at mindless dot com
//------------------------------------------------
Fernando <·······@must.die> writes:
> >4. Simple and fast GUI implementation for X-Windows.
> >Preferable NOT using Tcl/Tk.
> >What options are available. CLIM and CAPI
> >are not working on my system.
> >Liquid Common Lisp 5.0.3
>
> Amulet? I think that's the name of garnet's sucessor, but I'm
> not sure...
Yes, Amulet is/was the successor to Garnet, but Amulet is based on
C++, so it probably won't be of much help...
One could use Garnet, though...
--
Pierre Mai <····@acm.org> PGP and GPG keys at your nearest Keyserver
"One smaller motivation which, in part, stems from altruism is Microsoft-
bashing." [Microsoft memo, see http://www.opensource.org/halloween1.html]
From: Tim Bradshaw
Subject: Re: [Q] Some real application : help wanted
Date:
Message-ID: <ey31z5o2v0v.fsf@cley.com>
* Sandeep Koranne wrote:
> 4. Simple and fast GUI implementation for X-Windows.
> Preferable NOT using Tcl/Tk.
> What options are available. CLIM and CAPI
> are not working on my system.
> Liquid Common Lisp 5.0.3
> 5. I also need some info on "deploying" applications.
> I have Harlequin Liquid Commom Lisp 5.0.3. -> Primary development
> and hopefully deployment also.
I'm unfamiliar with the current liquid / hcl versions, but they
definitely support CLIM. If you're willing to think in its terms,
then CLIM is probably a fairly good match.
--tim
From: William Deakin
Subject: Re: [Q] Some real application : help wanted
Date:
Message-ID: <38C4EBCD.493C6DD3@pindar.com>
Sandeep Koranne wrote:
> 2. Numerical Algorithms in Lisp (Matrix operations like EigenValue)
Have you found anything?
www.cs.cmu.edu/afs/cs/project/ai-repository/ai/lang/lisp/code has some
numerical algorithms. I've also got some other bits and pieces from
othere places if you're intereste. Also I've written some qr and lu
decomposition code (which is very alpha at the moment) which I want to
work up into general eigenvalue stuff.
If you have found anything like this, I would be very interested if you
could point me in the right direction (I am mildly allergic to
reinventing the wheel). If not, I would be very interested in
developing some numerical eigenvalue stuff.
Best Regards,
:) will
From: Reini Urban
Subject: Re: [Q] Some real application : help wanted
Date:
Message-ID: <38c83bd4.47515774@judy>
William Deakin wrote:
>Also I've written some qr and lu
>decomposition code (which is very alpha at the moment) which I want to
>work up into general eigenvalue stuff.
>
>If you have found anything like this, I would be very interested if you
>could point me in the right direction (I am mildly allergic to
>reinventing the wheel). If not, I would be very interested in
>developing some numerical eigenvalue stuff.
xlisp-stat has all this kind of numerical stuff.
qr, lu, eigenvalue
--
Reini Urban
http://xarch.tu-graz.ac.at/autocad/news/faq/autolisp.html
From: Will Deakin
Subject: Re: [Q] Some real application : help wanted
Date:
Message-ID: <8aeknv$pn0$1@epos.tesco.net>
Reini Urban wrote
>xlisp-stat has all this kind of numerical stuff.
>qr, lu, eigenvalue
Thanks for the lead. I take it that this is the same xlisp-stat at
www.stat.ucla.edu/develop/lisp/xlisp/xlisp-stat?
Best Regards,
:) will
Sandeep Koranne <·······@natlab.research.philips.com> writes:
> 1. File parsing : recursive-descent would do
> ( I mean , if LALR generator is not there, then I
> will do recursive-descent). The type of files I would
> read are VHDL, Verilog and stuff.
> PS: Yes I know EDIF would be best :)
For most simple non-user stuff, I mostly do recursive-descent by hand
(non-user because it's easy to accept too much with r-d parsers, see
recent thread on comp.compilers).
For stuff where a parser generator is useful, there's Zebu (IIRC, then
the most current version can be found somewhere on Digitool's ftp
site, ftp://ftp.digitool.com/).
> 4. Simple and fast GUI implementation for X-Windows.
> Preferable NOT using Tcl/Tk.
> What options are available. CLIM and CAPI
> are not working on my system.
> Liquid Common Lisp 5.0.3
[snip]
> I think that CMU-CL Python is _quite_ good, and it runs well.
> So I am concentrating on CMUCL for now : I always have Liquid
> Common Lisp for really BIG applications.
I think Fernando D. Mato Mira is working on merging CLM and
CMUCL/Motif (also named CLM in some contexts) and possibly porting
Gina (CLM GUI-Builder). So maybe going this direction as a GUI
toolkit might sensible. Since CLM and CMUCL/Motif are quite similar,
writing stuff now using CMUCL/Motif and CLX and later porting to an
upcoming merged CLM should be feasible. While CLM is somewhat
lowlevel, given a couple of nice macros this can feel quite similar to
e.g. CAPI.
Regs, Pierre.
--
Pierre Mai <····@acm.org> PGP and GPG keys at your nearest Keyserver
"One smaller motivation which, in part, stems from altruism is Microsoft-
bashing." [Microsoft memo, see http://www.opensource.org/halloween1.html]
Hello,
I have downloaded and installed (succesfully) Garnet from CMU.
It looks good: I am starting to write the GUI frontend using
gadgets, for fast low level drawing I will use CLX.
I hope to be able to report what my experience was .....
Thanx for the suggestions: I am downloading the material and
reading the docs and playing around with it.
Sandeep
--
"Lets Make Things Better"
ED&T Test, WAY 3.37
Philips Research Laboratories Botstraat, 7
5656 AA, Eindhoven 5654 NL, Eindhoven
The Netherlands The Netherlands
Phone: +31-(040)-27 45250 +31-(040)-2573492
Fax : +31-(040)-27 44626
E-mail: ···············@philips.com
Hi Sandeep,
for some of the numerical algorithms have a look at xlisp-stat, which
includes a lot of this kind of code.
http://www.stat.umn.edu/~luke/xls/xlsinfo/xlsinfo.html
Marc
--
------------------------------------------------------------------------------
email: marc dot hoffmann at users dot whh dot wau dot nl
------------------------------------------------------------------------------
Forgot to tell you, but have a look at electric:
http://www.electriceditor.com/
a free gnu ecad program with a scheme extension module
Marc
--
------------------------------------------------------------------------------
email: marc dot hoffmann at users dot whh dot wau dot nl
------------------------------------------------------------------------------
Sandeep Koranne wrote:
>
>
> 1. File parsing : recursive-descent would do
> ( I mean , if LALR generator is not there, then I
> will do recursive-descent). The type of files I would
> read are VHDL, Verilog and stuff.
> PS: Yes I know EDIF would be best :)
>
Sorry this took so long
Harlequin Lispworks has a parser generator. I don't know the theory or
acronyms mentioned but:
<quote from LispWorks User Guide>
13.1 Introduction
The parser generator generates an LALR parser from a specification of a
grammar. The parser generator has a simple facility for the static
resolution
of ambiguity in the grammar and supports an automatic run-time error
correction
mechanism as well as user-defined error correction. Semantic actions can
be included in the rules for the grammar by specifying Lisp forms to be
evaluated when reductions are performed.
</quote from LispWorks User Guide>
Their fine manuals are online too:
http://www.xanalys.com/software_tools/reference/documentation.html
Hope this helps.
Jukka Kakko
--
Q: How far does an apple fall from a tree ?
A: One Newton.
--- Asta Salmi
On Mon, 6 Mar 2000 10:08:29 GMT, Sandeep Koranne
<·······@natlab.research.philips.com> wrote:
> Hello,
> I am in the process of writing some VLSI CAD applications
> in Lisp.
>
> I would like to integrate them all to form a meaningful
> software, for analysis of circuits and VLSI layouts.
>
> I am looking for some pointers (code, description, doc)
> for the following areas :
>
> 1. File parsing : recursive-descent would do
> ( I mean , if LALR generator is not there, then I
> will do recursive-descent). The type of files I would
> read are VHDL, Verilog and stuff.
> PS: Yes I know EDIF would be best :)
As Jukka Kakko pointed out, LispWorks comes with a parser-generator.
> 4. Simple and fast GUI implementation for X-Windows.
> Preferable NOT using Tcl/Tk.
> What options are available. CLIM and CAPI
> are not working on my system.
> Liquid Common Lisp 5.0.3
CLIM is an add-on product for LCL (Liquid Common Lisp), but CAPI is bundled
with LCL 5.
Contact ··········@xanalys.com or ············@xanalys.com for whichever kind
of info you want.
__Jason