that no nets will be checked for antenna gate and diffusion area,
and no antenna properties will be output to the LEF file. This can
greatly speed up LEF output file generation for a large design.
Thank you to Tamas Hubai for the code patch.
report it after "Failure to read in entire sub-tree". This will
not report every failing cell (since it quits reading after the
first failure) but will avoid the existing issue of printing
nothing and leaving the user with no feedback as to which cell
was the problem.
previously, MASTERSLICE layers would not be added to obstruction
layers made by "-hide". However, an nwell, for example, that
stuck outside of a prBoundary *would* be recorded, which was
inconsistent. Resolved this by allowing MASTERSLICE layers
in the OBS block, but only for layers that are not a substrate
type. NOTE: It may be better to just insist that a MASTERSLICE
layer define an obstruction type in the "lef" section of the
tech file, and treat it like routing obstructions. Alternatively,
one may question whether special obstruction types are needed at
all, as one could simply define an obstruction as a type without
a port label.
did not need to be changed, but the line below it that depended
on the layer being a contact type was wrong; the contact had
to be identified independently of the last layer type, which
might not be the contact cut type.
non-default rule. The code was first failing to identify the via
cut type from the generated via record, and then it was failing
to return to the non-default rule width after the route exits the
via. Both issues have been fixed. Thanks to Sylvain Munaut for
providing a reproducible test case.
new port labels are created for an existing port, then they must
take the existing port number. The code was previously causing
collisions between port numbers on different pins.
without SelectRootDef being set, which causes SelectClear() to
return without doing anything. However---question is why
SelectClear() needs SelectRootDef to be non-NULL since it is
clearing the selection, not the selection source. For now, just
patching the failing case, not trying to mess with the whole
selection mechanism.
lefLayer's "via" record even when the layer might not be a via,
causing potential issues with uninitialized variables. Not sure
if this is related to the bug that started this investigation,
but it was the only thing that looked relevant.
no way to implement boolean operators on labels, so any "label"
statement in the section can apply only to one magic layer. This
is regularly violated in most (all?) techfiles (due mainly to lack
of explanation and guidance). The addition of the "no-reconnect-
labels" option for cifinput made it worse, as it can cause a label
to be attached to the wrong layer and be stuck that way. Even
without the option, an attachment to a non-connecting type is a
problem; DIFF cannot simultaneously have a connection to both
ndiff and pdiff, so it will be one or the other, and the one not
connected can easily get labels moved to other nets. To avoid
this: (1) removed the "no-reconnect-labels" option, and (2) made
the automatic label reconnection smarter, as well as splitting it
into two different behaviors based on whether a label is being
created or manipulated from the command line (more or less the
original behavior) vs. being read from GDS or LEF. The new rules
assume that labels attached to a GDS type will all map to the
same plane in magic. To avoid excessive error messages from
existing tech files, a warning is issued only if "labels" changes
the plane of the target layer (a realistic solution rather than
the preferred one). Also: Fixed an error that causes a crash on
the "wizard" command "*watch" if the cell being observed is
read-only (see github issue #271).
in DEF read and write. The NONDEFAULT LAYER WIREEXT was assumed
to refer to the default wire extension at segments, when instead
it refers to the wire extension only at vias. The wire
extension at segments is presumably defined by the nondefault
segment (and wire extension at vias remains unimplemented, which
is probably not a big issue because everyone puts the wire
extensions into the via definitions anyway).
output as blockages. That allows obstruction layers to be placed
in a layout specifically for the purpose of being output as a
blockage. Otherwise, an obstruction layer is generally considered
non-electrical and will not show up as a node in the extraction
file, so obstruction layers were being missed entirely by "def
write". Also: Where "def write" complains about floating labels,
made an exception for labels on non-electrical layers (like
"comment").
when a label has no area and magic has to go searching for the
label area. The resulting behavior is better but is not really
a substitute for adding ports on the proper layers in the proper
locations for routing connections.
(and similar) independent routed segments in the SPECIALNETS section
(which was already working correctly for regular NETS). This solves
github issue tracker issue #264 from Christian Haufe.
Lookup() is used for command-line parsing in magic and accepts any
unique string that matches a partial keyword. LEF and DEF do not
allow partial keywords, so only LookupFull() is appropriate to use
for LEF and DEF file parsing. This fixes issue #263 from Christian
Haufe.
warning messages about ports being electrically connected when
those ports have names that match under rules of case-insensitivity,
and the .ext file is being read for the purpose of generating a
SPICE netlist, which is case-insensitive. Also: Corrected a crash
condition when using "extract path <name>" when directory <name>
does not exist.
nets to be avoided for running antenna gate and diffusion area
checks when doing "lef write", which is to check if the pin of
the net is flagged with use "power" or use "ground". This avoids
the need to use the (recently added) "lef nocheck" option (although
that still exists as an additional way to control which nets do and
do not get checked).
to create a list of net names to ignore for antenna gate and
diffusion area checks. This allows the nets not to have to be
selected in their entirity but selected by chunk only. This
reduces the time to write LEF on a large layout back to approximately
what it was before the change to include the hidden area from "-hide"
in the antenna area checks. Plus, it greatly reduces the time to
generate LEF for large layouts when not using the "-hide" option.
gate and diffusion area on each pin is done before erasing parts
of the cell that are to be hidden by obstruction layers. That
preserves the antenna information even when "-hide" is used. This
corrects the issue raised by Kareem Farid in the github issue
tracker #236.
previous behavior that had inadvertently been changed. In recent
versions, "load <absolute_path> -dereference" would incorrectly
apply the dereferencing to <absolute_path> rather than just its
subcells. Cleaned up the code around DBCellRead() in the process,
so everything is more straightforward (although probably more
could be done in that regard).
dereferencing, and making the behavior of "load" on the command
line (i.e., loading a cell from a file) the same as the
behavior of loading a cell as a result of expanding an unloaded
instance. In both cases, if "load -dereference" is used, and
a cell does not exist in any search path but does exist in the
original location, without dereferencing, then the cell will be
loaded from the original location. Also: Corrected an error
that has existed since adding the capability to read compressed
files, which causes magic to crash when attempting to run the
"crash recover" command (because that routine was mixing
compressed and regular file stream calls).
for compatible layout over the area of the pin rectangle in the
entire hierarchy of the cell, not just in the top level. This
corrects issues where pins are placed in the top level cell with
no metal underneath. It is written in such a way that it will
work regardless of whether paint is split across the hierarchy,
or the label spans different types (such as crossing a contact).
in the input. The name of one was being modified but was being
right-justified into the string, resulting in a name starting with
many space characters. Also: Fixed an issue with extresist where
a label that ends up with an empty string can become a node with
an empty string name in the .res.ext file output.
pin on a compatible area due to an incorrect type mask. (Also
note that this routine only works (after correction) for RECT
statements, but there should be a similar check for POLYGON
statements.)
contacts) take default values from the DRC section. Since both
are in lambda, but the DRC section uses a two-part integer and
modulus representation, if default values are taken before
scaling, the LEF layers may get rounded values. This has been
solved by marking values with -1 to indicate that they require
defaults, and then set those defaults (from scaled DRC rules)
after scaling all other tech values.
redundant (same name, different net). Previously, the method was
to keep the first such node and ignore all others except to add
their resistance and capacitance to the original node. This
prevented routines like "def write" from enumerating all nets
unless they had unique names. The new method keeps the additional
records including the node location where they can be found by
EFNodeVisit(), but flags them with EF_UNIQUE_NODE so that routines
like ext2spice or ext2sim can choose to ignore them. This implies
that this method could be used to reimplement "extract unique"
within "ext2spice" or "ext2sim" without altering label text. This
has not yet been implemented.
Both were corrected with respect to the definition of non-default
(taper) rules. "def write" was additionally modified to avoid
redundantly processing tiles where tile areas were merged together
to form a complete wire. There is plenty of room for optimization,
but the output appears to be matching the layout. Also: Revised
the definition of "(not) visible layers" to include labels attached
to those layers, so that turning off visibility of any layer will
also hide all labels attached to that layer.
option. This is important as the default units of 1nm are not
necessarily able to be converted to integer values if the
minimum manufacturing grid is an odd number of nanometers.
"slivers"; the error tended to produce artifacts (extra metal)
around contacts. Fixed an issue that caused the DEF write routine
to open the same file twice instead of a new file for the second
part of the DEF data, and then potentially hit a runaway condition
when trying to merge the two files together.
routine; that is incorrect, and it should have been fgets().
When dbFgets() is recast to a zlib version, then its use in
"def write" causes magic to crash.
attempt is made to write an abstract view to GDS. This behavior can
be overridded with the new command option "gds abstract [enable|disable]".
Also: Corrected extraction to allow split tiles to be set as the
reference tile for a node. Previously this was allowed only if the
tile was the first to be searched, but that can cause different tiles to
be marked as the reference depending on where the search starts,
resulting in different names for the same node in .ext files, which is
bad. Also: Modified the LEF annotation to avoid bad entries in the LEF
that would create layers in the layout where none exist.
This commit makes the code (mostly) C99-compatible, enabling to compile
it without the -Wno-error=implicit-function-declaration flag. This
way, Magic becomes usable on arm64 architectures, specifically on Apple
computers with M1/M2 SoC.
flattening based on cells which have a property "flatten". Also:
Modified the DEF read and write to convert DIEAREA into a
FIXED_BBOX property. This solves issues with placing of components
from DEF, when those components may not have have come from place &
route and may not have P&R bounding boxes. Also: Fixed the
documentation for the "dump" command, which was missing the optional
orientation in the description.
that are unattached (type = space) will cause magic to crash on
the command "lef write -hide"---The lowest level issue was in
SelectChunk(), so for good measure SelectChunk() now checks for
type == TT_SPACE, and "lef write" does also.
non-default rules and so can accurately capture wire widths other
than the technology LEF defaults (also corrected a bug with non-
default rules for "def read"). Corrected via handling in "def
write" to function as claimed (although the algorithm is still
naive and expects all contacts to be rectangular, which is usually
true but doesn't have to be).
with class, use, and shape information from the DEF file PINS record.
This is similar to what is done with LEF file annotation, but less
sophisticated (only erases an existing pin if it is an exact match
for pin name and location).
DEF reading to use vector fonts on PIN labels, with some ad hoc
rules for size and rotation (may need refinement). Modified the
DEF annotation (def read -annotate) so that the preferred position
of labels is on a wire leading out from a pin connection, which
is a "safer" place to put it, in case the layout was manually
edited between the DEF read and annotation. Fixed another two
crash conditions related to read-only views. Corrected a startup
error caused by an uninitialized variable used by the "wiring"
section of the tech file.
to be annotated from a DEF file. This is particularly useful if
"def read" has been used without the "-labels" option, and the
labels are needed at some point.
in github issue #149. This causes magic to no longer write log files
for "make" and "make install" but will properly exit with a non-zero
return code on any error during compile.
Also: Corrected the command "tech drc surround <type1> <type2>" so
that it now returns the correct value when <type1> and <type2> are in
the same plane. Added new command "tech drc directional <type1> <type2>"
which works the same way as "tech drc surround" except for directional
surround rules. Used this to generate vias from "def write" with the
correct metal surround amounts included in the via definiton. The
route analysis then ignores tile slivers that make up the surrounding
material around contacts. Also implemented a method that handles
routes that are made of multiple thin tiles due to the maximum horizontal
stripes rule. Now magic handles "def write" well except for not dealing
with non-minimum-width routes unless they're specifically called out as
"special" nets.
the label to be attached to a non-contact type when the label is
actually over a contact are not handled correctly when checking if
multiple labels should belong to the same port record.
due to issues of declaring global variables; rather than track
down the correct use, just reworked it so that the value is just
an integer and takes -1 as the default (fixed timestamping
disabled).
"gds datestamp" for LEF reads. When set to non-default, all cell
definitions created from LEF macros will be given the specified
timestamp, which allows the stamps of abstract views to match the
stamps of full views, preventing issues of updated timestamps
whenever a layout is switched between views.
record to the label structure to hold the port number. One major
issue stemming from this was reported in github issue #203 by Anton
Blanchard. This commit fixes that error.
client data generated by ext2spice and attached to a node's
nodeClient record; there is an initNodeClient() routine but no
corresponding freeNodeClient() routine. Eventually had to add a
callback function passed to EFDone() and EFFlatDone() to clean up
these entries. After doing that, valgrind reports clean for all
memory allocated within ext2spice (there are other things that are
not freed but not related to a specific command, so do not need to
be treated as leaks).
this limited ports to 16384, which seemed reasonable at the time.
However, the sky130_sram_macro layouts connect power and ground in a
way that when coupled with "extract unique" can generate tens of
thousands of ports and overrun the bit field, showing that automation
can do the unexpected. The solution was to split out the port number
from the label record as its own 32-bit value.
statements, with all "hard" connections being enumerated in the
same PORT entry, and "soft" connections (same label on unconnected
areas; e.g., through substrate or resistor device) being
enumerated as separate PORT entries, per the LEF spec. Also
corrected behavior with respect to the "lef write -toplayer"
option, which was treating each port label independently, and so
generating entries for lower layers of a port if there were ports
on those layers, in contravention to the "-toplayer" option.
Also: Added the PINS section to the "def write" output; this had
been left as a "to be completed" item but was never done in spite
of being easy to add.
commands on selections in a cell that is not editable. Moves
and Copies were already handled correctly; this correction fixes
Delete and transforms (e.g., rotates and flips).
an optional extra argument to the "select" command that can be used
to select labels by glob-style matching; e.g., "select area labels
VSS*" or "select less area labels *_1". This will help in managing
labels after flattening a standard cell design; e.g., by using
"select less area labels */VDD".
all geometry will be handled; the previous behavior expected only
a single rectangle per pin and so would only acknowledge the last
entry in any list of rectangles for the pin.
argument to "lef write" take an optional value which is a setback
distance, similar to "-hide", but specifically for limiting the
distance that pins can extend into the center of a macro.
to stop the search whenever a cell is not found. Used this to implement
a new option for GDS writes, "gds undefined allow|disallow" (default
"disallow") controls whether or not GDS with undefined references will
be allowed to be written. Similarly affects CIF and LEF writes, extraction,
and DRC (when running "drc check" from the top).
I'm seeing warnings when parsing a DEF with BLOCKAGES in it:
DEF read, Line 572201 (Message): Unknown keyword "BLOCKAGES" in DEF file; ignoring.
DEF read, Line 572202 (Message): Unknown keyword "-" in DEF file; ignoring.
DEF read, Line 572205 (Error): END statement out of context.
Skip over the section.
Corrected the "lef write -hide" command option so that obstructions
outside of the boundary are included in the obstruction list, in
addition to the block inside. This had previously been done
correctly for use with the "setback" option but would fail only
for setback = 0.
Ghazy) to read pin uses from a DEF file, including the three types
that are not defined by the LEF format (go figure). Expanded the
bitmask of label flags to include the additional use types. Also
shifted the label flag bitmask up to make additional room for more
port number, as there is no point in having unused bits in the
bitmask.
defMakeInverseLayerMap() from returning complete contact types when used
by the lefWrite command when writing LEF macros, which contain cut layers
but not entire contacts.
on cells with multiple ports per pin. As written, the code was
erasing all labels before creating a new port label, which would
erase all previous port labels. This should have been done only on
the first port.
read". While "lef read" normally annotates existing layout, this
option ensures that no additional cells are created from macros in
the input LEF file. (2) Added a check on the "Input off lambda grid"
warning during CIF/GDS input such that it is not repeated once issued,
as it tends to be output many times when it occurs.
are considered an exception to the "-toplayer" restriction; this is
because masterslice well/substrate layers will affect the electrical
connectivity between port and sustrate or well.
command when annotating an existing layout from a LEF database, if
there is a port in the layout that is shadowed by a label with the
same name that is not a port.
that produces a result that looks like "lef write -hide" in the middle
but "lef write" around the edge. Can be useful for catching all the
detail around the edges but obscuring/simplifying the bulk of the cell
interior.
Tim Edwards
Alessandro De Laurenzis
Anton Blanchard
Maximo
Dan Moore