I forgot to say that the port installs the "magic.1" man page; this
isn't a real conflict with magic(5), but could be a bit confusing;
please let me know if it is worth to change the man page name (is
something like magic-vlsi more appropriate?)
On 07/03/2019 06:12, Alessandro DE LAURENZIS wrote:
Dear ports@ readers,
this is a proposal for a new port: magic 8.1.224, a venerable VLSI
layout tool, written in the 1980's at Berkeley by John Ousterhout (now
famous primarily for Tcl), still very popular with universities and
small companies, and an integral part of Qflow (not yet in the tree),
where it is used as layout viewer and for parasitics extraction and DRC
(but it can do a lot more).
DESCR
=====
Magic is an interactive system for creating and modifying VLSI
circuit layouts. It is used to design basic cells and to combine
them hierarchically into large structures.
Magic understands quite a bit about the nature of circuits. It has
built-in knowledge of layout rules; during editing, it continuously
checks for rule violations. Magic also knows about connectivity and
transistors, and contains a built-in hierarchical circuit extractor.
It has a plow operation that permits to stretch or compact cells.
Lastly, Magic has routing tools to make the circuit interconnections.
Magic is based on the Mead-Conway style of design: it uses simplified
design rules and circuit structures that make it easier layout drawing
and permit Magic to provide powerful assistance, at the cost of
slightly less dense circuits.
There are 2 branches, distribution (8.1) and development (8.2); the
former is now reasonably stable, and should be expected to work without
undue issues; let's stay on that.
Upstream configure script is a wrapper around the one actually generated
by autoconf, so I used:
CONFIGURE_STYLE = simple
and patched it adding -std=c89 and -Wno-parentheses to CFLAGS, in order
to get rid of the huge amount of Wimplicit-function-declaration and
Wno-parentheses warnings.
Upstream makefile has been patched in order to log errors and warnings
to stdout.
Other patches have essentially to do with termios functions; in
particular, txGetTermState is not defined is OpenBSD; please carefully
review what I made in patches/patch-textio_txInput_c, since I just tried
to adapt some code found on the Internet used for similar situations;
neither I really understand if those functions are actually needed when
the Tcl/Tk version is used (this is the case for our port).
Taking MAINTAINER, tarball attached.
--
Alessandro DE LAURENZIS
[mailto:[email protected]]
Web: http://www.atlantide.t28.net
LinkedIn: https://www.linkedin.com/in/delaurenzis/
