* Bug: In AutoContactTurn::updateTopology(), in some case we were making
doglegs on segments in non-preferred routing direction (mostly M2).
This must be never be done. As, by construction, at least one segment
of the turn must be in the preferred routing direction, always use
this one to make the dogleg.
This bug is likely to explain the remaining "bad spin" in the
final self-check of the routing data-structure.
The decoupling of the cell gauge and the routing gauge implies that
the METAL2 & METAL3 terminals of macro blocks cannot be aligned on
the routing tracks anymore. That is, an horizontal METAL2 terminal
will not be on a track axis, but offgrid, so we no longer can use
a METAL2 horizontal segment to connect to it. Making an adjustement
between the offgrid terminal and the on-grid segment has proven
too complex and generating difficult configuration for the router.
Moreover, METLA2 terminal could be fully inside a METAL2 blockage.
So now, when the gauges are decoupled, we connect the METAL2 and
METAL3 the same way we do for METAL1: *from above* in the perpandicular
direction and using a *sliding* VIA. We assume that those kind of
terminals in upper metals are quite long.
* New: In Hurricane::Rectilinear, export the isNonRectangle() method
to the Python interface.
* New: In CRL::RoutingGauge, add function isSuperPitched() with the
associated boolean attribute. Set to true when each pitch of
each layer is independant (not low fractional multiples).
* New: In AnabaticEngine, add the ability to temporarily disable the
canonize() operation (mainly used in dogleg creation).
* New: In AutoSegment::canonize(), do nothing if the operation is
disabled by AnabaticEngine.
* Bug: In Session::_revalidateTopology(), disable the canonization
during the topology updating of a net. Too early canonization
was occuring in makeDogleg() leading to incoherencies when
performing the later canonization stage over the complete net.
Mostly occured in the initial build stage of the net.
* New: In GCell, add function postGlobalAnnotate(), if a layer
is fully blocked (above 0.9), typically, under a blockage,
add a further capacity decrease of 2 on the edges. So we may
handle a modicum of doglegs.
* Bug; In GCell::addBlockage(), removeContact(), removeHSegment()
and removeVSegment(), forgot to set the Invalidated flag.
This may have lead to innacurate densities.
* Change: In GCell::updateDensity(), more complex setting of the
GoStraight flag. This flag is now set if we don't have two
*contiguous* below 60% of density. We need free contiguous
layers to make doglegs.
* New: In NetBuilder, now manage a current state flag along
with the state flag of the *source* GCell. This flag is used
to tell if the GCell needs it's *global* routing to be done
using the upper layers (METAL4 & METAL5) instead of the
lower ones.
* New: In NetBuilder::setStartHook(), set the state flag of the
GCell to ToUpperRouting when processing a global routing
articulation and one of the base layer is obstructed
above 0.9.
In GCell with terminals, also set ToUpperRouting when there
are some in METAL2 / METAL3 and the gauge is not super-pitched.
* New: In NetBuilder, function isInsideBlockage(), to check if a
terminal is completely or partially enclosed in a blockage.
* Change: In NetBuilderHV::doRp_AutoContact(), remove support for
trying to put on grid misaligned METAL2/METAL3.
Instead systematically access them from above.
Do not cover with fixed protection terminals that are already
enclosed in blockages.
* Bug: In NetBuilderHV::doRp_AutoContact(), always add the terminal
contact in the requested GCell and not the target/source one,
in case the terminal span several GCells.
* Change: In NetBuilderHV::doRp_Access(), create the local wiring
according to the RoutingPad layer.
* Change: In NetBuilderHV::_do_xG(), _do_2G(),
create the global wiring in upper layers, according to the
ToUpperRouting flag.
* Change: In NetBuilderHV::_do_xG_xM3(), now delegate to
_do_xG_xM3_baseRouting() and _do_xG_xM3_upperRouting() if the
density at terminal level is above 0.5.
* New: NetBuilderHV::_do_xG_xM3_baseRouting() and
_do_xG_xM3_upperRouting() separated function to manage the
local routing.
* Change: In NetBuilder::_do_globalSegment(), if the currently
processed GCell or it's source is in ToUpperRouting mode,
move up the global segment. Do *not* use the moveUp() function
which would create doglegs unwanted at this stage.
* New: In KatanaEngine::annotateGlobalGraph(), call postGlobalAnnotate()
on the GCell after the blockages have been taken into accound to
add the penalty.
* Bug: In Track::getPrevious(), correctly manage the 0 value for the
index argument. Strange it didn't show earlier.
Same goes for Track::expandFreeInterval().
After a Cell has been created in memory (by parsers or Python scripts)
we can annotate it with the Spice parser so it will know the right
order with which to create the subcircuit call ('x').
* New: In CRL::Spice::load(), add support to read the ".subckt" card
and guess the right ordering for generating the 'x' (subcircuit
card call).
* Bug: In Spice::SpiceBit & Spice::BitExtension, when a Net bit property
is removed, if it's the currently cached property in BitExtension
it may lead to a crash. So when a property is destroyed, we must
also clear the cache (see remove(), clearCache() & onReleasedby()).
I'm wary that this could also happen on other kind of cached
extensions...
* New: In CRL::NamingScheme, new method vhdlToVlog() to translate back
VHDL net name into Verilog. Currently only changes "()" into "[]".
Used to generate the commented SPICE interface for Alliance compliance.
* Change: In Spice::Entity, previously all the ordering where removed
between each run of the SPICE parser, but the orders read from
SPICE file (mostly standard cells) must be kept. So add a flag
ReferenceCell to prevent the removal by ::destroyAll().
* Bug: In NetlistModel::setCell() template, forgot to send the
"layoutAboutToBeChanged" signal before disconnecting the
netlist datas. This seemed to go unnoticed until ported
under Qt5. Was generating strange random core dump when
running the engines with a net kept selected. Disgraceful
crash during demos...
* Bug: In Selector::_preDestroy(), *do not* iterate over the _cellWidgets
map as we destroy it's elements! We end up in destroyed ones...
Instead destroy the first one until the map is empty.
* Change: In CellWidget::_redraw(), the boundaries, that is, instances
abutment boxes where drawn *after* everything else, so they where
hiding more interesting informations (especially at low zoom level)
like routing wires.
Now draw them *first* so they are *below*.
* Change: In CellImage::setScreenCellWidget, fully copy the display
settings from the reference widget (not just part of them).
* Bug: In CellImage::toImage(), if the abutment box of the displayed
cell is not *fully* includer in the viewer area, do *not* try to
reframe it. The reframe() method seems to be buggy, works OK
just without calling it.
* Change: In unicorn/python/unicornInit.py, when iterating over all the
loaded modules, now we have "namespace" that have a __file__
attribute which is set to None. Skip them.
* Change: In stratus1/st_net.py, "is" must be replaced by "==" when
doing string comparison.
* Change: In stratus1/st_parser.py, open file in "rb" mode instead
of "r" (so we get a bytestream as now required by Python).
* Bug: In Viewer::SelectionPopup(), the window attribute
Qt::WA_DeleteOnClose was *not* cleared. So the window was deleted
after first use while it was though staying allocated.
Again, generating weird crashes.
Took the occasion to slightly redesign the behavior to select
and highlight individual components.
* Bug: In Isobar3::PyTypeManager, the accessors _getCppTypeName() and
_getPyTypeName() where returning string *by value*, hence, short
lived copies.
But, in _setupPyType() and _addToModule(), as we interface with
the Python/C API, we extract the c_str(). Which where removed as
we used temporary objects. Leading to memory corruption and weird
crashes.
Now returns "const string&" so the c_str() stays allocated.
* Bug: In CellWidget::DrawingQuery::drawGo(), the display condition was
wrong, it was requiring *both* width & height to be above the display
threshold. Either one is sufficient to activate the display.
Was causing the selective diseapearance of gates at low zoom level
and printed version.
* Change: In CellPrinter, force the display threshold to one pixel in it's
internal CellWidget used for printing.
* Bug: In Katana::DataNegociate::update(), when computing the length of
source & target extension on a perpandicular segment, must use the
extensions of the *directly* connected AutoSegment (the baseSegment)
and not the canonical one that may be different, so with unrelated
extensions.
* Bug: In cumulus/plugins.block.HTree._rrouteHTree(), the RoutingPads
for the input and output of the buffer where sometimes put too
close from each others, giving the pitch of the vertical tracks.
Now shift one pitch left the vertical branchs of the H-Tree.
* Bug: In cumulus/plugins.block.HTree._rrouteHTree(), also shift down
one pitch the horizontal branch, due to track rounding they *may*
end up on the same track, generating a short.
* Bug: In Anabatic::layerAssign(), during the step of desaturation of
GCells that contains too much terminals (i.e. lot of *local*
congestion), the desaturation threshold was har-coded to 8.
Wich is fine for symbolic cmos but way too low for Flexlib.
End result was that most straight wires where moved towards
the upper layers, creating congestion (imbalance of layer
densities).
Now the parameter:
* "anabatic.saturateRp" (default value:8) is correctly taken
into account.
* Fix: In Manipulator::avoidBlockage(), the dedicated function to check if
there is an obstacle in the way of a non-prefered routing wire (done
for metal2 connecting to terminals only) was too naive.
We were checking the tracks for obstacles crossing exactly the axis
of the segment. And for near-miss this is not enough. Now check on
the whole x-span to be used by the segment.
Code borrowed and simplified from Track::addOverlapcost().
* Change: In Manipulator::moveUp(), the default value for the extra
reservation allowing a long wire to move up need to be customized
for Flexlib/StdCellLib (allow successful routing of ChipFlow/MPW4).
Add two new configuration parameters to katana:
* "katana.longWireUpThresold1" : the length, expressed in number
of *slice height* above which a global wire is considered a
*long* wire (not close interconnect).
* "katana.longWireUpReserve1" : the extra number of free tracks
that must remains free in the up layer after the move up,
in each GCell traversed by the wire. Expressed in number of
tracks, but can be non-integer (float, for instance: 1.5).
* Bug: In Katana::PowerRailsPlane::Rail::doLayout(), to avaid tracks too
close to an offgrid obstacle in the preferred routing direction, we
expand the width/height of the segment by one pitch. BUT it seems to
still be too close for Flexlib and StdCellLib, so there is an ad-hoc
patch based on the *name* of the cell library. Update it to take
"StdCellLib" into account.
NOTE: This is likely to explain why we still got overlap in the
track coherency check in very rare occasions.
* Bug: In Katana::DataNegociate::update(), when computing the allowed
free interval for the segment axis deduced from the perpandicularly
connex segments, we account for the extension of the connecting
VIA. Those extension varies according to the kind of VIA and are
given by getExtensionCap().
We were accounting for the source & target extension VIA on the
parallel segments, assuming that source/target would not swap when
the perpandicular is moved. Which is *not* true.
Now account for the extension of the *connecting* VIA on all ends.
* Change: In AutoSegment::getTopologicalInfos(), enrich the list of
perpandicularly connected segment with wether they are connex by
their *source* or *target* contact. Mainly to be used by
DataNegociate::update().
* Change: In cumulus/plugins/core2chip, instead of the user providing
an explicit mapping towards the harness I/O pins, we expect that
the core block must have I/O pins with names matching thoses of
the harness. That way, connections are automatically made.
* Change: In cumulus/plugins/block/htree, if the root signal of the
H-Tree is a bit from a vector (like "io_in(0)"), then remove the
vector index notation on all the stem name of all the sub-nets.
Done by the unbitify() function: "io_in(0)" => "io_in_bit0".
* Change: In EtesianEngine::_postCreate(), issue a warning if the list of
feeds is empy (configuration: "cfg.etesian.feedNames").
* New: Add a FeedCeels::getFeedByWidth() method to get feeds by their
width in DbU::Unit and not only pitches.
* Change: In Placement::Slice::fillHole(), invert the tie filling and
feeds filling stage. Now we first try to fill the row hole with
feeds, using the widest first, and if they are not configured or
too wide, use the tie.
As the tie *should* also be integrated in the feed list, we may
suppress altogether the fallback tie filling step. Keep it for now.
* Change: In Placement::slice::createDiodeUnder(), the inserted diode
*may* be smaller than the feed it replace. So, in this case, add
a complementary feed to fill the gap.
NOTE 1: Out of lazyness, we add only *one* complementary filler
cell. So there *must* be one of a width wich correspond
exactly to the difference between the original feed and
the diode. Otherwise, gap will remains.
NOTE 2: With wider feed cells, they may cross the GCell border.
But we must insert the diode under the GCell, otherwise
the global routing will be defective. So, for now, reject
feeds that cross the boundary. Must be done more smartly
by inserting the diode over the left or right side of
the feed.
* Change: In AnabaticEngine::setupPrerouteds(), take into account the
number of Pins. Now consider a net containing multiple Pins and, at most,
one segment as *non-routed".
This case may specifically happens for nets with pins on the north
and east side, which are slightly *inside* the abutment box (to be
seen by the router) and draw with them their *outside* direct
connection wire.
* Bug: In cumulus/plugins.chip.CoreWire, no longer put the north or
east side external Pin *exacyly* on the abutment box but *one pitch*
inside so they are correctly seens by the P&R (must be *inside*
the area of a GCell).
It is now possible to automatically nest a core block inside a harness
frame, like we do for an ordinary chip whith I/O pads. The DEF harness
file "user_project_wrapper.def" must be made available though the block
configuration variable:
conf.cfg.harness.path = './user_project_wrapper.def'
A first small example is given in:
alliance-check-toolkit/benchs/counter/sky130_c4m
The harness layout is stripped from it's native power grid (but keep
the power ring). I/O pad information in block/configuration is
slightly "bent* to manage pins instead of complete I/O pads.
* Bug: In cumulus/plugins.block.Block.setupAb(), the routingBb was not
set up when working in chip mode. Now set (to the corona AB).
* Change: In cumulus/plugins.chip.__init__, move there the CoreWire
class (from chip/pads.py) so it can be shared with the harness
version of pads.py.
* Change: In cumulus/plugins.chip.powerplane, compute the intersection
between the vertical supply stripes and the deep horizontal power
lines in a smarter fashion, so two (or more) vertically contiguous
BigVias are merged into one (two BigVia side by side where causing
mimimal spacing distance violation on the cut in Sky130).
* New: In Anabatic::AutoSegment::create(), allow the created segment to
be in any supported routing layer, and not only the bottom H & V.
Modifications impact the *two* overload of the function.
* Change: In Anabatic::NetBuilderHV::doRp_AutoContacts(), for punctual
METAL1, the protection has to be in METAL2. Bump the layer depth
to correctly use the updated verstion of AutoSegment::create().
* Change: In AnabaticEngine::checkPlacement(), for the Pin, check that
it's layer is in the routing gauge before anything else.
* New: In Katana::NegociateWindow::createTrackSegment(), if the track
nearest the segment axis (refTrack) do not exists, call a
breakpoint just before crashing.
Jean-Paul Chaput
gatecat