cb1653b157Found that I needed to apply the same correction as the last commit to the "tagged" GDS/CIF operator, or else this will have exactly the same issue as "bloat-all".
master
8.3.635
magic-8.3
R. Timothy Edwards
2026-04-10 13:39:17 -0400
a3f5e665d1Found an extremely bad error, where the "bloat-all" GDS output operator does a second search to reset tiles and clear the "processed" mark. This second search was only searching on the types being bloated *into* and not the original types being bloated. This caused tiles to remain marked as "processed" which could in turn cause the tiles to be ignored forever after, potentially resulting in bad GDS output.
8.3.634
R. Timothy Edwards
2026-04-09 11:13:30 -0400
2eea849c06Corrected a crash condition in R-C extraction (extresist) caused by old code that still references devices like FETs assuming four ports. I resolved the crash condition but have not worked on fixing the code to properly handle a variable number of terminals per device.
8.3.633
R. Timothy Edwards
2026-04-07 21:15:56 -0400
6e295d030eAlso updated the HTML command reference for the "expand" and "unexpand" commands to include the options just implemented.
8.3.632
R. Timothy Edwards
2026-04-04 21:03:06 -0400
ceba050a21Fixed the mess that is the "expand"/"unexpand" command set. This had numerous problems, the main one being that each of the three commands was inconsistent: "expand toggle" inverts the expansion of the selection and syncs with the layout. But "expand" expands the layout where instances overlap the box and does not sync to the selection, and "unexpand" unexpands the layout where the box completely surrounds instances. Added a set of options to "expand" and "unexpand" so that these functions can be made consistent with each other. All varieties of the function now always sync the selection and the layout.
R. Timothy Edwards
2026-04-04 20:47:04 -0400
f7cceed5e3Made a fairly major overhaul to the toolkit for generated devices. The main changes are: (1) Changed the hash function used to generate the 6-character suffix for generated device cell names. The original hash function is not good for ensuring unique names, and can cause cell name collisions (two different parameter sets having the same cell name). The chance of name collisions should now be diminishingly small. (2) Modified the string passed to the hash: (A) Ordered the parameter names alphabetically, since iterating through dictionary keys is not guaranteed to be in any specific order, leading to different strings for the same parameter set (B) Normalized numerical parameters, so that "2", "2.0", "2e0" are all hashed the same, again to avoid having multiple cell names for the same set of parameters. (3) Fixed a problem in which when changing parameters for a cell instance, the instance would become unselected and the instance name would be lost and revert to magic's auto-generated name. (4) Fixed the annoyance of having a pop-up dialog whenever magic decides that a parameterized cell name is not being used anywhere, and it can safely delete the cell. (5) Fixed an issue where the check for whether a cell can be deleted is not run consistently. The result is, I hope, a much more pleasant experience with generated cells.
8.3.631
R. Timothy Edwards
2026-04-02 17:13:17 -0400
5d35ae38b5Corrected an issue causing MASKHINTS values to fail to be written back to the file, instead resulting in a property key only and no value; this in turn produces an error when read back in. Another cycle of reading and writing causes the property to disappear altogether.
R. Timothy Edwards
2026-04-02 14:45:08 -0400
05561b90f3Solved three issues: (1) There was still one place that dimensional units were not being printed, which was "setlabel", with both "setlabel size" and "setlabel offset". This has been fixed. (2) Fixing (1) surfaced an error in the text helper dialog, which was not setting units properly when checking the size and offset entries. (3) Fixed a very long-standing problem in which port labels were showing up with the "pale" style of a subcell's label being drawn on top when the parent and child had the same label in exactly the same position. This was due to a parent-first search resulting in the child cell being visited last, so its "pale" style label being drawn last. Added a flag for label searches for display to reverse the search order, visiting children first and then the parent, so that the top cell's labels are the last to be drawn.
8.3.630
R. Timothy Edwards
2026-04-01 21:17:54 -0400
06eab7feb6Removed three lines of code that were meant to be a test of how much performance can be sped up if the DRC process can be interrupted at much finer intervals. Unfortunately, the method cannot work without an additional method to ensure that magic is immediately aware when a search routine has been entered recursively and to then break out of the search and clean up after itself. For now, I am just commenting out the code so that magic isn't unstable.
8.3.629
R. Timothy Edwards
2026-03-26 19:44:00 -0400
9ade4c931eUpdated the version to go along with the merge of two pull requests from Darryl Miles. The second of these should finally put to rest all the compile-time issues related to "termio/termios" (yay).
R. Timothy Edwards
2026-03-26 09:23:58 -0400
1bb5316d8dutils/magsgtty.h removal and fixup
Darryl L. Miles
2025-07-31 22:59:00 +0100
2f26237b8bGHA: main.yml/main-aarch64.yml: WASM build
Darryl L. Miles
2026-03-25 12:39:20 +0000
003506ba62CmdLQ.c: fix MAGIC_WRAPPER code section(s) for WASM build
Darryl L. Miles
2026-03-25 12:39:20 +0000
d659b5e1dftextio.h: add include <stdarg.h> due to va_list type use in file
Darryl L. Miles
2026-03-25 12:39:20 +0000
a0afe242c8extract/ExtNghbors.c: fix older compiler support for label
Darryl L. Miles
2026-03-23 22:41:59 +0000
88ca77f6a4GHA: main.yml: remove vezzal task
Darryl L. Miles
2026-03-25 12:39:20 +0000
9d967bdf64GHA: canary-matrix.yml: apt-get update
Darryl L. Miles
2026-03-23 22:40:51 +0000
7fd2ef8100I admit I was juggling with too many balls in the air yesterday and managed to commit some code that I thought I had tested--- but apparently what I tested was not the latest version of the code. So 8.3.627 is quite broken and won't even compile. This update fixes that.
8.3.628
R. Timothy Edwards
2026-03-25 14:07:54 -0400
c862a0ddfdGHA: main.yml/main-aarch64.yml: WASM build
Darryl L. Miles
2026-03-25 12:39:20 +0000
efee2ffe89drc/DRCbasic.c: fix: syntax error
Darryl L. Miles
2026-03-25 12:39:20 +0000
864360a7a3utils/magsgtty.h removal and fixup
Darryl L. Miles
2025-07-31 22:59:00 +0100
a8edc04b5cCmdLQ.c: fix MAGIC_WRAPPER code section(s) for WASM build
Darryl L. Miles
2026-03-25 12:39:20 +0000
131d86f998drc/drc.h: fix: syntax error
Darryl L. Miles
2026-03-25 12:39:20 +0000
dd890ca1a1textio.h: add include <stdarg.h> due to va_list type use in file
Darryl L. Miles
2026-03-25 12:39:20 +0000
27fbaeed17extract/ExtNghbors.c: fix older compiler support for label
Darryl L. Miles
2026-03-23 22:41:59 +0000
7d11f875c5GHA: main.yml: remove vezzal task
Darryl L. Miles
2026-03-25 12:39:20 +0000
4a4cd34677GHA: canary-matrix.yml: apt-get update
Darryl L. Miles
2026-03-23 22:40:51 +0000
e789f18523Yet another pass at the problem caused by implementing DRC exceptions; avoided the use of "signed char" altogether by just making the drcc_exception value a bit mask with the lower 7 bits being the index and the upper bit being the flag for exception (0) vs. exemption (1), with the value 0xff being reserved for "none" (no exception or exemption). This implementation should avoid any issues caused by ambiguity in the use of signed vs. unsigned char.
8.3.627
R. Timothy Edwards
2026-03-24 09:37:50 -0400
460a357730Corrected what appears to be a long-standing error in the GDS read routine, in which it calls the same "cleanup" routine as for CIF files, which in turn processes the top level cell. However, GDS doesn't have the same concept of a top level cell, so it ends up re-processing part of the last cell read. This would appear to be harmless, except that there is some kind of scaling discrepancy in which the cell bounding box can get incorrectly scaled by the cleanup routine. The solution appears to be simply (?) to not do the "top level" processing. Bypassing the CIFPaintCurrent() from the cleanup routine when reading a GDS file appears to solve the problem.
8.3.626
R. Timothy Edwards
2026-03-22 13:11:27 -0400
7ceeccef56Corrected an issue found by Harald Pretl and solved by Claude. Use of "char" does *not* implicitly mean "signed" on ARM architectures.
R. Timothy Edwards
2026-03-22 11:14:19 -0400
66faf1d907Added documentation for the attribute labels in the "label" command description, because otherwise it only appears in the tech file reference, where it doesn't really belong since the labels are a user-level feature, even if they do affect extraction. The new feature of using gate attributes for parameters was added to the command description. I noticed that the tech file reference made mention of a gate label in the form of ":ext <name>=<value>", but this seems not to be implemented anywhere. At any rate, the method just committed supercedes that.
8.3.625
R. Timothy Edwards
2026-03-18 19:28:23 -0400
37db9e453bExtended the use of attribute labels to allow gate attributes to specify additional parameters for a device. Any gate attribute (label ending with "^") that has an "=" in it is considered a parameter, and is output with the device in the .ext file. This gives a way to add specific parameters to a device, which otherwise have no meaning to the layout editor. Note that prior to this extension, the gate attribute was used only to set the device index. If a gate attribute is made for a device index, it cannot have "=" in the name, so it remains backwardly compatible. Only non-parameter attributes are passed on to ext2spice.
R. Timothy Edwards
2026-03-18 16:41:42 -0400
2d5c4be6ddAdded a few lines of code to the toolkit management to work around the issue of a device name in a netlist having different case than the procedures defined for the device in the PDK. The difference in case is transparent to SPICE, so it should also be transparent to the PDK.
R. Timothy Edwards
2026-03-18 12:18:42 -0400
a9673e45aeAdded inductors to the list of devices for which the toolkit can call a device generator for an ideal component in a netlist.
8.3.624
R. Timothy Edwards
2026-03-17 11:42:16 -0400
8f95efc257Added a feature to the toolkit scripts for generated devices to allow ideal devices in an input netlist to be mapped to automatically generated cells. This works if the PDK toolkit defines a device called either "capacitor" or "resistor". The device must be defined such that it can determine sane parameters for the drawn device from only a value given in the netlist (capacitance in farads, or resistance in ohms).
R. Timothy Edwards
2026-03-17 11:32:53 -0400
fd12c39c37Modified the toolkit scripts so that the "add_dependency" procedure actually works, and also can be called multiple times. The chaining of commands and bindings had earlier been done for one of the widget types (selection) but not the others (entry or checkbox).
8.3.623
R. Timothy Edwards
2026-03-16 20:02:09 -0400
7d9167257aCorrected an error caused by a recent commit that was attempting to avoid bogus values for array pitch in the direction in which a cell is not arrayed: If the array declares an ANGLE of 90 or 270, then rows and columns are effectively swapped for the purpose of figuring out which coordinate direction should be ignored in the input. The original example did not use ANGLE and so the exception was not handled.
R. Timothy Edwards
2026-03-16 12:50:17 -0400
058b320c3fAlso applied the same correction as the previous commit, to hierarchical extraction (as opposed to flat extraction, which is what the previous commit fixed).
8.3.621
R. Timothy Edwards
2026-03-15 15:55:48 -0400
fee4b887c0Corrected an issue in ext2spice in which devices extracted as ideal, unmodeled capacitors or resistors are missing a space between the 2nd terminal and the device value in the output netlist. This affects the ability to produce correct output for, for example, a metal fingered capacitor using the "device" property to generate an ideal capacitor in the netlist.
R. Timothy Edwards
2026-03-15 15:40:10 -0400
04e91d640dCorrected a problem in which if a label exists that is only connected to material in a child cell, and the label is a point or line label, and the parent cell has no material other than the label in the immediate area, then the "interaction area" will have zero area, which will cause problems as area checks on a zero-sized area don't do anything useful. The solution is to ensure that interaction areas do not have zero area, but will expand around point labels. This should not have any other implications, because the interaction area is just a way to limit the amount of processing; larger interaction areas do not affect the extraction result (except in this case, where it fixes the error).
8.3.620
R. Timothy Edwards
2026-03-12 11:50:21 -0400
7d9210a3f1Updated the HTML reference for the "property" command, which did not describe the "list" option (which was just as well, since the option didn't work).
8.3.619
R. Timothy Edwards
2026-03-11 16:43:29 -0400
03bbc544b2Fixed a long-standing error that had gone unnoticed that prevents the use of "property list <key>" to return NULL if <key> is undefined, instead of printing an error message that cannot be suppressed. Scripts which wish to test whether or not a bounding box exists will of course want the "quiet" version of the command.
R. Timothy Edwards
2026-03-11 16:35:52 -0400
ee79bba5e4Decided to change the name of the new cifinput/cifoutput operator from "labeled" to "tagged", since there is already an operator called "labels", which is confusing.
R. Timothy Edwards
2026-03-11 15:52:42 -0400
15943d0cb1Implemented new cifinput/cifoutput operator "labeled" which operates like "bloat-all" except starting with a label and expanding into a type rather than starting with a layer. This is equivalent to what many tools refer to as "stamping".
R. Timothy Edwards
2026-03-11 15:06:24 -0400
7bdd9e1d4fReworked the method from the previous set of commits, which creates yet another property type "plane". This allows mask hints to be handled like tile planes. That is, after all, what they represent. Although this change is transparent to the end user (apart from a slight performance improvement that is probably not noticeable), it allows for a number of useful future extensions, including the ability to handle non-Manhattan geometry in mask hints, and the ability to view and edit mask hints like any other layout.
8.3.618
R. Timothy Edwards
2026-03-10 14:25:02 -0400
725c8e9235Made a few corrections to yesterday's code. Most are minor, but one can potentially cause a crash condition.
8.3.617
R. Timothy Edwards
2026-03-09 14:40:53 -0400
73ffad3802Implemented the new feature discussed in the github discussion #492. This features allows two new instructions in the tech file "drc" section: "exception <name>|none" and "exemption <name>|none". The purpose of these rules is to allow DRC rule exceptions for certain areas which are marked, usually by a GDS identifier layer. Because GDS identifier layers can be cast as "mask-hint" properties, the DRC rule exceptions make use of that (recent) feature. So rules can be specified as "exceptions" (rule applies for layout inside the marked area) or "exemptions" (rule is ignored for layout inside the marked area). Rules following the statements will have the exception or exemption applied until another exception or exemption statement is given. Either "exception none" or "exemption none" will stop applying any rule exception or exemption. This is especially useful for applications like SRAM where there may be a number of rule exceptions on different layers. The previous way to handle that was to create new layers in magic for any layer that needed a rule exception. That mainly resulted in very messy tech files, and a large number of defined layers. This implementation is both simpler and cleaner.
8.3.616
R. Timothy Edwards
2026-03-08 20:12:58 -0400
037daf1121Found an issue with coupling capacitance where contact types not surrounded by metal are not checked for perimeter coupling parasitic capacitance to other layers.
8.3.615
R. Timothy Edwards
2026-03-05 18:04:53 -0500
d0ef32de0fTracked down a DRC issue found recently where spurious errors get drawn outside the area being worked on. Tracked it down to a missing check for an invalid rectangle. Fixed now.
R. Timothy Edwards
2026-03-05 10:17:29 -0500
0d9b862c0eCorrected a minor error in ext2hier.c where views with no devices and no ports are still supposed to be output as instances if they have been marked as abstract views. Such views were getting output as cell definitions but not instanced, which although is irrelevant for simulation, would potentially cause LVS errors.
R. Timothy Edwards
2026-03-05 09:35:28 -0500
71dc472797Modified the toolkit code to make use of the "units" command; otherwise, the existing code would produce results that vary by units, with potentially incorrect label sizes and spacing between instances when generating layout devices and subcells from a netlist.
8.3.614
R. Timothy Edwards
2026-03-04 13:34:31 -0500
2929ef583eMinor correction to the new "repeat <steps> ... endrepeat" function for tech files, as the first implementation was repeating one too many times. Also: The wrong type was being read for the number of steps, although the resulting value was correct.
8.3.613
R. Timothy Edwards
2026-03-02 12:24:46 -0500
ecd6ec56aeModified the routine for the cif-output "bloat-all <dist>" operator to make it slightly more efficient. Noting, however, that it is still highly inefficient relative to the stepped "grow" + "and" (or "and-not") operators that it was supposed to replace, I also implemented a convenience function "repeat <steps>" ... "endrepeat" to the tech file syntax. This allows the "grow" + "and" series to be defined with a few lines, whereas if the steps have to be spelled out, the series can be scores or hundreds of lines long.
8.3.612
R. Timothy Edwards
2026-03-01 21:35:00 -0500
f3478cba7bFixed a few issues related to the "save <file>.tcl": (1) Dimension properties were not output correctly. (2) Magic insisted that the output file was <cell>.tcl, not <file>.tcl, although the error was only in the messaging (3) Magic was incorrectly renaming the cell to <cell>.tcl after writing the file.
R. Timothy Edwards
2026-03-01 11:29:25 -0500
ba5154698dThought better of the recent change to make "snap" no longer change units; although that is a more proper usage, it also happens to break a lot of existing scripts. Since virtually all of these scripts start out with "snap internal" to force units, I have kept this single usage, when "units" is still set to "default". That keeps the older scripts from breaking, while hopefully scripts get updated over time to use "units" instead of "snap". Note that there are still ways to make this behavior break, such as running "units microns" to set units and then sourcing a script that assumes that "snap internal" will change the units, which will then not happen. But batch scripts should be handled property with backwards- compatibility mode.
8.3.611
R. Timothy Edwards
2026-02-28 20:07:00 -0500
1023461ca5Minor change to the "save <file>.tcl" method to only paint from TECHDEPBASE, as the layers below that cannot be directly painted; this avoids putting things like "paint checkpaint" into the output, which are harmless but useless.
8.3.610
R. Timothy Edwards
2026-02-27 12:08:35 -0500
197763e46eModified the output of "save <cell>.tcl" so that (1) it strips the ".tcl" off of the name, so that the cell name remains the same, and (2) it uses "units" instead of "snap", as the latter is no longer used for setting the unit type.
R. Timothy Edwards
2026-02-27 11:18:27 -0500
cb7855235aAdded a new cifinput operator "not-square" that is useful for differentiating between bar and square contacts on input. Also: added handling of the "mask-hints" operator for GDS input as well as GDS output. This is a bit more realistic now that the mask hint properties are handled as scalable integers and not as character strings.
8.3.609
R. Timothy Edwards
2026-02-26 12:48:00 -0500
8f684ad5beFound and fixed the actual error underlying the problem with grid rescaling while reading a GDS AREF record. The previous commit is still valid and prevents excessive warnings and unnecessary rescaling when a bogus value is given for an array pitch when the array has only one row or one column. But the scaling was still incorrect and would potentially cause an error if an array pitch was at a finer scale than any data read before it, forcing a valid grid rescaling. This now works correctly.
8.3.608
R. Timothy Edwards
2026-02-25 09:59:28 -0500
f3404f67b7Came across a weird error today in which some tool inserted bogus data into an array reference in GDS; This may be ignored by most tools (which are not trying to keep track of grid scaling) which read the bogus value which then gets discarded because it is unused (X pitch is irrelevant when there is only one column, and Y pitch is irrelevant when there is only one row). But since magic will try to interpret all input relative to the grid, a bogus value can cause serious problems, and magic needs to actively ignore the appropriate input in the case of rows = 1 or columns = 1.
8.3.607
R. Timothy Edwards
2026-02-24 20:56:27 -0500
cedd64adcbQuick fix to the script use of "units", which needs to expand the list argument if the units were saved as a variable.
R. Timothy Edwards
2026-02-24 16:40:07 -0500
afca58f162Fixed a number of places in the Tcl code where "snap" had been used to force internal units to be used; this usage is now deprecated and needed to be changed to the "units" command. Also: Fixed a separate issue with getting both "view bbox" and "view get" to return values in the currently selected units. Also: Fixed an issue with "property" when setting a property to a list passed as a single value (e.g., "property FIXED_BBOX [box values]"), which needed to null the string after each value.
R. Timothy Edwards
2026-02-24 15:59:21 -0500
95baea1c22Corrected another place where the property handling had not been converted to the new PropertyRecord type. Since this was in the CIFPropRecordPath() subroutine, which has always been problematic and generally useless, I took the opportunity to make that function dependent on an option set by "gds savepaths true", with the option set to "false" by default, so "path" properties will no longer be created from GDS input unless the option is manually set to "true". Also: Found that the returned boolean from "box exists" is sometimes getting overwritten with a NULL result, causing a spurious Tcl error messages to pop up. I could not find where this overwriting happens (probably in a tag callback), so I just took the hack approach of converting a NULL result to "false", which is always safe to do.
8.3.606
R. Timothy Edwards
2026-02-19 11:40:46 -0500
751757a02cAnother correction to one function call that was not updated for the new property structure, causing the "instance abutment" command to potentially generate an incorrect set of values.
R. Timothy Edwards
2026-02-19 10:27:03 -0500
c32bae1a24Corrected a minor error from yesterday's commit that will cause magic to complain about reading GDS_START and GDS_END properties from a .mag file, although its fallback method handles the properties correctly, anyway, so there is no actual impact other than the error message.
R. Timothy Edwards
2026-02-19 09:42:41 -0500
0c913eca59Another correction to a code error from a commit earlier today. Fixes an infinite loop when querying properties during CIF read.
8.3.605
R. Timothy Edwards
2026-02-18 20:27:33 -0500
2f00f6d8f1Corrected an error in the output of string properties when writing .mag files caused by the last commit---Failure to put a space between the key and the value. Fixed.
R. Timothy Edwards
2026-02-18 19:15:14 -0500
9ab7b77dc4Thought better about suddenly changing the .mag file format with new properties, so created a "property compat" setting and made it TRUE by default. This makes magic print all properties as type "string" on output into a .mag file. Which is fine, since it converts all values to the right type on input anyway. The only thing that backwards-compatibility mode prevents is user- defined properties that are not strings. That is a very rare case and can be done by turning off comptability mode. Some time in the future compatibility mode can be changed to be default false, but there's probably no real need to do so.
R. Timothy Edwards
2026-02-18 16:17:03 -0500
00c0208f18Corrected an error in which the "global substrate node" flag does not get transferred to a new node during a node merge; consequently, the identity of the default substrate node gets lost, and node capacitance to substrate ends up being placed on node "0"--- Probably okay for simulation, but not the expected result.
R. Timothy Edwards
2026-02-18 12:28:48 -0500
cb30ac369bExtended the "property" command and modified the way that properties are handled. Properties were previously only character strings, which had become cumbersome because properties were being used for mask hints and bounding boxes, with the necessity of constantly converting values from string to integer and back, which can cause a performance impact as well as just being messy. The main difference to the command is the addition of an optional first keyword argument for the property type, which can be "string", "integer", "dimension", or "double". All types except "string" can consist of multiple values. Multiple values can be specified as separate arguments on the command line, so that, for example, values of FIXED_BBOX or MASKHINTS_* no longer need to be quoted. In addition, this completes the handling of "units" implemented recently, as all properties of the type "dimension" can be entered in the current units, will display in the current units, and will scale with the database.
R. Timothy Edwards
2026-02-18 10:48:47 -0500
7e9b6fb61eFixed the logic for determining which of two node names is "best". The code was not enforcing an order of precedence, which meant that the lexigraphical order of EFHNBest() was able to override the simpler case of one of the node names being a top level port.
8.3.604
R. Timothy Edwards
2026-02-16 12:02:32 -0500
3b1de8ff40Corrected an egregious error made when converting "extresist" from reading the .sim file format to reading the .ext file format, which is that .sim files use capacitance values of fF while .ext files use capacitance values of (nominally) aF. So there was a conversion factor of 1000 left over from the .sim file reading code that needed to be removed, or else parasitic values come out to be 1000 times too high.
8.3.603
R. Timothy Edwards
2026-02-06 10:50:29 -0500
10a6c8635fCorrected an error with the new "extresist" code which fails to set the location of a node which was previously read as a port. But the port location was not set when reading the port, only the drivepoint, so the location is undefined and can cause a program crash.
8.3.602
R. Timothy Edwards
2026-02-05 17:57:43 -0500
e2292f5d70Modified additional code in ext2spice in the subcircuit instance output which matches the code for the subcircuit definition output, so that the pin order matches between the two. The previous fix prevented pins from going missing in the subcircuit definition, but didn't match that change in the instances. This should resolve github issue #488.
R. Timothy Edwards
2026-02-05 17:17:25 -0500
736c507fe8Preparing to merge another round of development work on extresist; this work is incomplete but cleans up the code a lot and removes unused code and improves misleading variable and subroutine names, as well as correcting an issue with "extresist" not recognizing the "extract path" setting. There are no known impacts to the operation of extresist itself.
8.3.601
R. Timothy Edwards
2026-02-04 07:54:43 -0500
c20a267a2bUpdates to the README file for development work in progress.
R. Timothy Edwards
2026-02-03 21:09:05 -0500
bad0b67ce8Additional work to clean up the code and make some things less confusing. Saving work here while rebasing to changes in master.
R. Timothy Edwards
2026-02-03 17:30:19 -0500
727649b308Saving work so I can rebase on changes made to the magic extract code which are relevant and need to be included. Current state is that hierarchical extresist "basically works" but entry and exit points through the hierarchy are not being examined, so results are currently based on port positions and not actual connections. Also, proper distribution of coupling caps has not yet been worked on.
R. Timothy Edwards
2026-01-30 16:55:08 -0500
4dde62b206Start of project. Only some comments in a README file so far.
R. Timothy Edwards
2026-01-28 17:20:29 -0500
94edc2a23dCorrected an issue where "extresist" crashes magic if the .ext file it needs to read doesn't exist.
8.3.600
R. Timothy Edwards
2026-02-03 20:59:28 -0500
b248f186ecIdentified a potential issue where the terminal area and perimeter calculation could result in the wrong count of shared terminals if the device consists of more than one tile, and corrected it.
R. Timothy Edwards
2026-02-03 17:27:57 -0500
1656866f41Accidentally reduced by one the count of devices sharing a terminal area in a recent commit; fixed this.
R. Timothy Edwards
2026-02-03 16:14:08 -0500
99297e33ecModified the way that ext2spice determines which nodes are port nodes; this should fix issues with ports not appearing in a subcircuit's port list.
R. Timothy Edwards
2026-02-03 11:45:20 -0500
727833fcd3Made another update to go along with the previous commit, that changes the EFHNIsGlob() function to treat only names listed in the Tcl array variable "globals" as global, and ignore names that simply end in "!".
8.3.599
R. Timothy Edwards
2026-01-31 13:02:35 -0500
feb5d61294Removed two lines of code in the EFHNBest() routine that prefer a name with a trailing exclamation point over any other name. The handling of global names might be properly implementable, but this is not it, as it will favor a non-port default name or a subcircuit name over a port.
R. Timothy Edwards
2026-01-31 10:40:29 -0500
4b120eb417Corrected two errors related to extraction: (1) All parasitic extraction: The "defaultperimeter" and the "defaultsideoverlap" commands were failing to exempt types other than space from the list of edges from which fringing capacitance is evaluated. This led to incorrectly considering the boundary between types such as poly and nfet, or between metal1 and rm1, to be sidewall areas. The "default" statements are supposed to consider the most common usage, so the code has been changed to make sure that only edges from material to space are considered. In the rare case that a material-to-material edge in the same plane should be considered a sidewall, the non-default statements can be used instead. (2) Hierarchical parasitic extraction: Magic was incorrectly adding capacitances for subcells which had been output already when handling subcircuit connections during "ext2spice". This duplicate counting has been eliminated.
8.3.598
R. Timothy Edwards
2026-01-30 16:47:36 -0500
790f0196d4Fix termio.
Gon Solo
2026-01-23 16:18:10 +0100
55eadcfb90Updated some of the HTML command reference documentation to reflect changes just made in the "extract" and "extresist" commands.
8.3.597
R. Timothy Edwards
2026-01-28 17:05:00 -0500
b4f62abb40Removing my file of notes during development of the extresist code; the file exists in git history but shouldn't be kept in the current master branch.
R. Timothy Edwards
2026-01-28 15:59:08 -0500
deefe0e3a3Updated the version after merging the new extresist code changes.
R. Timothy Edwards
2026-01-28 14:35:45 -0500
57c33c48c7Final work on the first stage of the extresist overhaul. The "extresist" command continues to work as before. However, the method now reads from .ext files instead of .sim files, so generating ".sim" and ".nodes" files is no longer necessary. In addition, the core code of "extresist" was put directly into ExtCell.c so that full R-C extraction can be run using "extract do resistance" followed by "extract all", without needing to run "extresist" at all other than to set parameters (e.g., "extresist tolerance 10").
R. Timothy Edwards
2026-01-28 14:31:38 -0500
76f97c90e5Slowly working through things needed for removing the ".sim" file dependency of "extresist". Stopping and committing work in order to rebase from the master branch.
R. Timothy Edwards
2026-01-27 11:56:47 -0500
Jean-Paul Chaput
Oran C
Dan Moore
Darryl Miles
anirban
Gonsolo
R. Timothy Edwards
Torleif Skår