luke
/
magic
mirror of https://github.com/RTimothyEdwards/magic.git
1
0
Fork 0
Code Issues Packages Projects Releases Wiki Activity
677 Commits 6 Branches 731 Tags 36 MiB
Commit Graph

112 Commits

Author SHA1 Message Date
Tim Edwards f6c6e2c29c Reverted a change from a prior commit that removed backslashes in 
front of brackets in LEF names.  These have a special meaning
that is not yet handled correctly, but removing them is the wrong
approach.
 2020-06-05 14:55:45 -04:00
Tim Edwards 643bd096ea One more fix. . . 2020-05-29 16:51:20 -04:00
Tim Edwards 09e85dedab And the corresponding pointer. 2020-05-29 16:39:40 -04:00
Tim Edwards 112e58e371 And one more hash key type needed to be changed in the other call 
to HashInit().
 2020-05-29 16:36:07 -04:00
Tim Edwards 82106387a6 Corrected the type of hash key on the property definitions hash 
table in lefWrite.
 2020-05-29 16:26:37 -04:00
Tim Edwards 44ca5d996b Added handling of PROPERTYDEFINITIONS block in LEF output based on 
properties in the cells.
 2020-05-29 16:12:09 -04:00
Tim Edwards 4ad7ce3cf4 Modified LEF read and write to preserve macro PROPERTY lines. This 
probably needs revisiting, because "lef write" and "lef writeall"
need handling to generate the PROPERTYDEFINITIONS block for the
PROPERTY entries to be correct.
 2020-05-29 14:31:48 -04:00
Tim Edwards bad4842707 Corrected a improperly placed call to freeMagic() from the last 
commit.
 2020-05-29 13:10:03 -04:00
Tim Edwards 9372a9e0e8 Added more behavior to "lef read" annotation to include handling 
bracket characters in the LEF file that are escaped with backslashes.
 2020-05-29 12:51:21 -04:00
Tim Edwards a57f024587 Making the "lef read" usage for annotation of an existing layout 
somewhat flexible with regard to pin name translations between
the LEF and the layout.
 2020-05-29 12:37:10 -04:00
Tim Edwards bf061f0012 Additional change to readspice script to handle various issues with 
pins not matching between netlist and layout due to delimiter changes
or case sensitivity.
 2020-05-29 11:55:30 -04:00
Tim Edwards 36f2d0cca9 Corrected an out-of-order code block that can cause the USE statement 
to be printed twice for a pin during "lef write".
 2020-05-29 09:35:54 -04:00
Tim Edwards 483f15360a Added support for "PORT SHAPE" in LEF files. 2020-05-28 22:06:22 -04:00
Tim 'mithro' Ansell b47d4c5642 Removing extra `$Header$` 2020-05-27 20:35:27 -04:00
Tim Edwards e63e8f5761 Corrected the expected position of top and bottom layers in a 
DEF VIA ENCLOSURE when doing "def read".  Thanks to Ahmed Ghazy
for finding the error and providing the patch.
 2020-05-27 18:02:36 -04:00
Tim Edwards b6f4553cef One additional place where the port number needed to be taken from 
the existing label on "lef read", where the port record is never
read.
 2020-05-26 20:40:41 -04:00
Tim Edwards 8b159a07cd Corrected LEF annotation so that it does not try to change or any 
port indexes, or create duplicated port indexes.
 2020-05-26 17:58:05 -04:00
Tim Edwards 582539e95d Modified LEF and DEF writes to set VERSION to 5.7, as nothing in 
the output is known to be incompatible with LEF/DEF version 5.7.
 2020-05-26 17:25:44 -04:00
Tim Edwards da9c180670 Modified lefWrite with the odd trick of adding a zero to prevent 
fprint from printing "negative zeros", which routinely happens in
the ORIGIN output of LEF files with "lef write".
 2020-05-26 17:19:57 -04:00
Tim Edwards 84f63ad545 Corrected segfault condition when running "lef write" on a cell 
with split tiles (error caused by previous commit).
 2020-05-25 16:13:42 -04:00
Tim Edwards 994c6aaf57 Modified the LEF write routine so that it correctly handles contact 
layers (apart from the fact that contacts are output as magic's
contact layer representation, and not as cuts;  this still needs to
be handled properly).
 2020-05-25 12:26:01 -04:00
Tim Edwards 2af38a4191 Updates from running the fix-ending-whitespace script. 2020-05-23 17:13:14 -04:00
Tim 'mithro' Ansell c81d9add00 LEF writer: Fix indenting. 
I missed the LAYER value and the geometry was one level to high.

Previous;
```
    PORT
         LAYER li1 ;
      RECT 1.145000 1.075000 1.690000 1.275000 ;
      RECT 3.720000 1.075000 4.490000 1.275000 ;
         LAYER met1 ;
      RECT 1.105000 1.260000 1.395000 1.305000 ;
      RECT 3.765000 1.260000 4.055000 1.305000 ;
```

After;
```
    PORT
      LAYER li1 ;
        RECT 1.145000 1.075000 1.690000 1.275000 ;
        RECT 3.720000 1.075000 4.490000 1.275000 ;
      LAYER met1 ;
        RECT 1.105000 1.260000 1.395000 1.305000 ;
        RECT 3.765000 1.260000 4.055000 1.305000 ;
```
 2020-05-23 17:03:03 -04:00
Tim 'mithro' Ansell 2c066c8df0 Remove trailing whitespace. 2020-05-22 20:14:13 -04:00
Tim 'mithro' Ansell e2f0832cdf LEF writer: Move formatting properties to #defines 2020-05-22 20:14:13 -04:00
Tim 'mithro' Ansell c04a33cbf1 LEF writer: Use 6 decimal places of accuracy in output. 2020-05-22 20:14:13 -04:00
Tim Edwards 89c9335727 Modified area accumulation routines to avoid double-counting contacts 
during "lef write".
 2020-05-22 20:13:33 -04:00
Tim Edwards 77d17a4fa7 One correction to yesterday's commit for "lef write" enhancements: 
If the last pin to be parsed generated no output, then the
obstructions would generate no output as well.
 2020-05-22 16:14:27 -04:00
Tim Edwards 1e9334664c Modified the LEF write routine so that it will not output ports 
that have no geometry (that do not exist on planes defined in
LEF).
 2020-05-21 21:53:23 -04:00
Tim Edwards 6adb5dbacf Enhanced the "lef write" routine: (1) Calculates gate and diff 
areas and writes ANTENNAGATEAREA and ANTENNADIFFAREA values.
(2) Determines "USE POWER" or "USE GROUND" from label names
matching Tcl variables $VDD and $GND, if the USE has not been
registered as a cell property (knowning the use allows magic
to avoid writing an ANTENNADIFFAREA for power rails, although
doing so should not be an issue).
 2020-05-21 16:26:24 -04:00
Tim Edwards 6635817383 Modified LEF read to give all labels the sticky flag so that they 
cannot be moved from the layer on which they are defined in the
LEF file.
 2020-05-21 13:47:14 -04:00
Tim Edwards 838591bdf2 Corrected unfortunate typo from today's previous commit, and 
corrected a wrong typecasting that results in a compile warning.
 2020-05-10 20:01:05 -04:00
Tim Edwards f0d2c8db0f Corrected the LEF read routine so that it parses an ORIGIN line in 
a LEF macro that has parentheses around the coordinates.  Weirdly,
this is requires by the LEF/DEF spec, but is rarely if ever seen
in actual LEF files.  Go figure.
 2020-05-10 12:21:04 -04:00
Tim Edwards 7e1fcc8ef6 Fixed lefWrite, which was not including drc.h and so was messing up 
one of the arguments to the DRC technology query.  Also, drc/drc.h
itself did not export that particular function, so it was added.
 2020-04-07 20:08:42 -04:00
Tim Edwards 08fabeedd5 Merge branch 'master' into bplane 
Merging fix to "lef write" to preserve SITE, from master branch.
 2020-04-01 11:02:08 -04:00
Tim Edwards e98b768c45 Corrected "lef write" to add the SITE definition that it may have 
been preserving in the cell properties (which was inadvertently
omitted).
 2020-04-01 11:01:13 -04:00
Tim Edwards 706e342171 Merge branch 'master' into bplane 
Conflicts:
	lef/lefWrite.c

Merged change to "lef write -hide" method from the master branch.
 2020-04-01 10:46:06 -04:00
Tim Edwards d6cadeb0cd Decided that "lef write -hide" should always use widespacing rules, 
because otherwise all pins will flag metal-to-obstruction spacing
within the cell if the cell is wide enough that the obstruction
layer satisfies the width requirement for the rule.  It is too
complicated to try to find specific places where the wide spacing
might not be needed.  Potentially this could be a problem for
technologies that define a number of graded wide-spacing rules,
as the largest-width rule is always used now by "lef write -hide",
and the largest-width rule could theoretically allow enough space
to route through, which would cause a short that cannot be
detected.  That would be a pathological case that may not show up
in practice.
 2020-04-01 10:39:43 -04:00
Tim Edwards daf93e5bad Merge branch 'master' into bplane 
Conflicts:
	VERSION

Merged from master correction that should prevent "lef write" from
creating pins with no geometry in them.
 2020-04-01 09:01:33 -04:00
Tim Edwards e296fc5ba0 Corrected an error in lefWrite that should have avoided degenerate 
labels by expanding a zero area label rectangle, but then if "select
chunk" returns nothing, it sets the area to the zero area label
rectangle instead of the expanded one that it just created.  This
is the reason that "lef write" is producing pins with no geometry
in the LEF file output.
 2020-04-01 08:59:19 -04:00
Tim Edwards d742550edc Merge branch 'master' into bplane 
Conflicts:
	VERSION

Merge from master (LEF read:  Read all tokens for LEF CLASS record)
 2020-03-30 09:37:42 -04:00
Tim Edwards b5e06455c5 Corrected LEF read routine so as not to truncate LEF "CLASS" to a 
single token (LEF CLASS records may have a subclass, such as "PAD
INPUT").
 2020-03-30 09:36:24 -04:00
Tim Edwards 521efeb929 Corrected uninitialized variable problem in lefWrite. 2020-03-25 15:21:05 -04:00
Tim Edwards aab318176b Corrected lefWrite, which was not initializing the count of items 
written, so that a LEF output with no ports would fail to write
the OBS ... END around the obstruction geometry.
 2020-03-25 15:19:28 -04:00
Tim Edwards 79ada35815 Merge branch 'master' into bplane 
Conflicts:
	VERSION
	extract/ExtMain.c
	lef/lefWrite.c

Pulled master branch changes to lefWrite into the bplane branch.
 2020-03-25 11:23:16 -04:00
Tim Edwards 195e096d64 Modified lefWrite to ensure maximum spacing per all widespacing rules 
for any gap between a pin touching the cell boundary and the internal
obstruction layer.
 2020-03-25 11:07:54 -04:00
Tim Edwards 6bae7c25c4 Enhanced the "lef write -hide" command option to check for metal 
wide spacing rules; should result in LEF views that can import
back into magic without DRC errors.
 2020-03-25 09:29:16 -04:00
Tim Edwards 007806521c Updated the "lef write -hide" to understand and use widespacing rules 
(may still need additional enhancements, as it does not specifically
look for layer-to-obstruction rules).
 2020-03-25 09:25:32 -04:00
Tim Edwards ff0ba7f89d Merge branch 'master' into bplane 
Conflicts:
	VERSION

Merged recent changes from master back into bplane, as the efficiency of
bplane for doing extraction on large layouts is unquestionably better.
Fixed the implementation of DBMoveCell() for bplane.  Corrected an error
in the bplane version of dbScaleCell() that enumerates cell uses but
does not free the list.
 2020-03-22 14:07:25 -04:00
Tim Edwards 276bf1d78a Corrected a bad error from a few commits ago, probably 198, which 
can deallocate the cell name and causes all sorts of unexpected and
unwanted behavior.
 2020-03-21 12:40:35 -04:00