|
|
Magic Mailing List |
|
From: cfk (cfk AT pacbell DOT net) Date: Sun Mar 02 2003 - 15:32:12 EST
Dear Tim:
Thank you very much for adding a way to find the cursor in Magic units.
It allows me to navigate around the screen much better. I can execute the
command 'cursor' from the Tcl window and see the current location. I can now
appreciate the origin (little black rectangle on my screen), which I couldnt
do before.
I next uncommented the bind line #797 and did a fresh make config, make
tcl and make install-tcl. I was expecting that either I would have a window
somewhere, or I could spawn a window with 'o', or the first invocation of
'cursor' would add the cursor position to the title bar (or somewhere
visible), but alas, I havent yet figured out the magical invocation, but
I'll poke around a bit more.
Let me jump subjects for a bit. I have been recently studying the
standard cell library from http://www.rsl.ukans.edu/~mlinhart/magic and it
is very interesting. I can getcell his various cells like invf100.mag and
place them in a layout. Just for info, it appears that the cells are based
on the University of Miss. cell library, as doing a google search on a cell
name finds documentation at the UMiss web site that describes the cells a
bit.
I am currently trying to understand why when I place a cell such as the
simplest one INVF100.mag and add paint for just the Vdd or GND connection on
metal1 using the box (so its 8 units wide), I get three errors pointing me
at P-Well spacing of at least 9 and width of P-Well of at least 10. Thats
with a GND metal1 paint, I get a corresponding set of errors with a Vdd
metal1 paint for N-Well on the top of the cell. I had thought that with
Magic 7.1 yesterday, I was able to paint metal out from a cell for Vdd, GND
and signals without error. I suspect it really is something simple that I am
just missing as the width of metal1 and the width of either P-Well or N-Well
should have little to do with each other. I also thought I had spent a bit
of time routing some metal1 between some of his cells. I have to admit that
moving back and forth twixt magic 7.1 and magic 7.2 is a bit of a pain as
just doing a make clean (or even make veryclean) in the magic 7.2 directory
doesnt seem to remove all the pieces. I have been doing a 'ln -s magic-7.1.5
magic', 'rm magic', 'ln -s magic-7.2.27 magic' to try to switch back and
forth between magic souce directories much like kernel compilation, but
there are some pieces spread hither and dither that I dont understand yet.
You might comment on the proper procedure for a dual magic installation
consisten with your developement efforts that will allow someone to give you
comparative feedback as you develop 7.2X.
I have a comment on the compilation, its not a criticism, just a
comment. I noticed that when I compiled magic-7.2.27 that at the 'make tcl'
step, I now have a myriad of warnings from the compiler cc1 of the form
'warning: changing search order for system directory "/usr/include" as it
has already been specified as a non-system directory'. I know that you have
been working on solaris compilation issues and dont know if this one gives
you any pause or not. The resulting executable does create a reasonable
executable so its not broken, just a little different.
Again, I compliment you on your abilities as I would have a tough time
keeping up with the development of a major software program and also working
at the same time.
Charles Krinke
http://home.pacbell.net/cfk
cfk AT pacbell DOT net
|
|
|
|