* New: In Kite, in protectCagedTerminal(), check that the perpandicular
top METAL3 track is free before inserting the fixed protection.
They are added *after* the Pre-Routed stage and therefore we have
to check they do not cross.
* Change: In Kite, in NegociateWindow, print statistics before fixing
segment in the pre-routed stage to get correct results.
* Change, In CRL Core, in Measures when an addMeasure is performed
twice, let the new data erase the previous one (instead of keeping
the previous one).
* New: In Katabatic, in <AutoContact>, this class is no longer derived
from ExtentionGo. With the simplificated AutoContacts, there is no
reason to do so, and it will save some QuadTree insertions/deletions.
New factory function AutoContact::createFrom(Contact*) which try to
build an AutoContact on top of a Hurricane::Contact. Of course that
base contact *must fit* into one of the predefined Contact
configurations (Terminal, Turn, HTee or VTee).
NOTE: This implies that the pre-routed segments & contacts *are*
correctly articulated, which is not the case when a Cell is read
from disk in "ap" format. The pre-routing feature must be used for
now without any re-read from disk. We will implement a re-articulating
pre-process in the future.
* Change: In Katabatic, in <AutoContact> derived classes, the ::updateCache()
method now display an accurate error message if a segment is connected
but has no AutoSegment conterpart (i.e. the lookup fails).
* New: In Katabatic, in <AutoSegment>, the ::computeOptimal() method is
short-circuited for pre-routed segments, the optimal axis position is
considered to be the one it is currently on (i.e. we trust the designer).
* New: In Katabatic, in <KatabaticEngine>, the ::loadGlobalRouting()
method now accept a map of excluded nets (same as Knik). This map is
the one of pre-routed nets.
* New: In Katabatic, in layer assignment, do not try to displace fixed
segments...
* New: In Katabatic, in <AutoSegment>, new flag SegUserDefined and related
methods to know if a segment comes from the global router (Knik) or
is pre-routed (supplied by the user).
* New: In Kite, In <BuildPowerRails>, support (exclusion) for pre-routed
nets.
* New: In Kite, In <GraphicKiteEngine> new menu entry for running the
router on pre-routed nets ("Detailed Pre-Route"), also integrated
in the all-on-one route command.
* New: In Kite, In KiteEngine, new method ::_initDataBase() that group
all the initialisation steps. It is a mix of calls between Knik and
Kite initializations which are intertwinneds (may have to devellop
a shared common base at a later point). It creates the Knik grid,
then the Katabatic grid, then load pre-routed wires and power rails
and protect isolated RoutingPads.
Add support for a map of pre-routed nets (to be excluded for
Knik calls).
The method "::run()" now uses function flags, firstly to know if
it is managing pre-routed wires or general purposes ones.
* New: In Kite, in <NegociateWindow>, the "::run()" methods has now two
modes. The normal one and the 'KtPreRoutedStage' that is for routing
pre-routed nets. When in pre-route stage, the wires are fixed at the
end of this step.
* New: In Kite, in <TrackElement> add decorator for AutoSegment
isUsedDefined().
* New: In Kite, in <TrackSegment>, the various ::canDogleg() methods
returns false for a pre-routed (user-defined segment).
* New: In Kite, in PyKiteEngine, added new method runNegociatePreRouted().
Placement management:
* Change: In <metis>, always disable the hMetis support regardless of
it being detected or not as the placer is still unable manage the
final bin contents.
Routing gauge management:
* Bug: In CRL Core, <vsclib/alliance.conf>, set the correct pitches and
size for the routing layers and the cell gauge.
* Change: In Katabatic & Kite, extract the correct extension cap for each
routing layer from the layers characteristics (cache then in
Katabatic::Configuration).
* Change: In Katabatic, <AutoSegment>, create segment with the wire width
defined in the gauge. For AutoSegment created on already existing
Segment from the global routing, adjust the width.
* Change: In Katabatic, <AutoSegment>, more accurate information about how
a segment is connected to terminal via source and/or target.
The flag SegStrongTerminal is splitted into SegSourceTerminal and
SegSourceTarget (but still used as a mask). So now we can know by
which end an AutoSegment is connected to a terminal.
* Change: In Katabatic, ::doRp_Access(), create constraint freeing segments
not only when HSmall but also when VSmall (more critical for <vsclib>).
Otherwise we may see AutoSegments with incompatible source/target
constraints.
* Change: In Kite, BuildPowerRails, do not create blockage on PinOnly
layers *but* still create power rails planes. This is a temporary
workaround for <vsclib> where the METAL1 blockages overlaps the
terminals (it was fine for Nero, but they shouldn't for Kite).
* Change: In Kite, <RoutingEvent>, if a TrackSegment is overconstrained,
directly bybass it's slackening state to DataNegociate::Slacken.
Also rename the flag "_canHandleConstraints" to "_overConstrained",
seems clearer to me.
Miscellaneous:
* Change: In CRL Core, <Utilities>, add a "pass-though" capability on the
mstream to temporarily make them print everything.
Miscellaneous:
* Change: In <crlcore>, in display.conf use the same display threshold
for both METAL2 & METAL3.
In alliance.conf, the side of VIAs in the gauge is 2l (not 3l).
In kite.conf, separate edge densities for H/V.
* Change: In <Cell>, in flattenNets() use flag as argument, not a
boolean. Do not create rings for clock or supply nets.
* Change: In <DeepNet>, in _createRoutingPads() do not create rings
for clock or supply net (duplicated policy as in Cell::flattenNets()).
* Bug: In <ControllerWidget>, at last find the bad signal disconnect
that was causing ungraceful messages.
* Change: In <knik>, in Edge display occupancy/capacity in the string
name. Improved display progress and debugging capabilities.
Improved exception catch & breakpoint managment:
* Bug: In <PaletteWidget>, in updateExtensions() replace the calls to
deleteLayer() by delete. This cause the widget to be immediatly
erased instead of waiting for the event queue to be completly
processed. This was causing the widget to be left in a incoherent
state when stoping at a breakpoint.
* Bug: In <BreakpointWidget>, in execNoModal(), flush the main event
loop (QApplication::flush()) *before* lauching the *local* event
loop. This is to ensure all widgets are in their final state when
waiting (especially <PaletteWidget>).
* Change: In <ExceptionWidget>, new method catchAllWrapper() to
execute any std::function< void() > function/method with a "try"/
"catch" wraparound and lauch the widget in case something is catch.
* New: In <hurricane>, support for a oberver pattern, backported from
<katabatic> with an Obervable capable of being linked to any
number of Obervers.
* New: In <Cell>, made it observable to detect Cell change, currently
emit two kind of signals:
- Cell::CellAboutToChange : *before* any change.
- Cell::CellChanged : *after* the change has been completed.
* New: In <UpdateSession>, in Go::invalidate() add the Cell owning the
Go to the UPDATOR_STACK (of course the cell is added only once).
In addition, when the Cell is added, send a notification of
Cell::CellAboutToChange to all it's observers. The slave instances
are also invalidated.
Conversely in UpdateSession::_preDestroy() for each invalidated
Cell send a Cell::CellChanged notification to all observer.
The UPDATOR_STACK has been slightly amended to accept Cell which
are not Gos. Prior to this, the Cell where completly excluded from
the UpdateSession mechanism, so it's instances where never actualised
of anything referring to the Cell for that matter.
Note: we use two different mechanisms to transmit a Cell change,
observers and the slave instance map. I think at some point it
should be unificated.
* Change: In <CellViewer>, make it a Cell observer to redraw when the
cell is modificated (also update the palette).
Uses the catchAllWrapper() to protect all critical actions.
* Change: In <GraphicTool>, no longer need of cellPreModificated and
cellPostModificated signals. Now done through the Cell obersvers.
* Change: In <mauka>, <etesian> & <kite> now uses the catchAllWrapper
method for protection (need to split methods in two, to be able
to pass it as argument). No longer emit cellPreModificated and
cellPostModificated.
Support for RoutingGauge in P&R:
* Bug: In <placeandroute.py>, the connection from the internal power
ring to the connectors was not done correctly. Wrong contact layers
leading to a gap.
* Change: In <BuildPowerRails>, detection of the corona signals based
on how the "pck_px" pad is connected. No longer based on name
matching.
* Change: In <placeandroute.py>, support for 2 routing metal only
(3 metal in the technology).
* Change: In <katabatic> & <kite> support for a "top layer" limitation
on the routing gauge, this allows to use only two routing metals
(METAL2 & METAL3). Work in progress.
- New: Added FreeBSD/Ubuntu patches from Otacilio De Araujo
(<otaciliodearaujo@gmail.com>).
* ./kite:
- New: Makes use of cbug to display the (scary) debug messages.
- Change: Re-enable deletion of Knik in Kite destructor.
- New: In NegociateWindow/RoutingEvent, adds a more comprehensive stage
"Repair". Perform in three stage: first try to place with a relaxed
constraint (one GCell on each side). Second try to minimize the faulty
segment. Third perform another "repack perpandicular" but this time
the faulty segment is re-inserted *before* any of it's perpandiculars.
- New: In RoutingEvent::cacheAxisHint(), when a segment has a parent, that
is comes for a "moveUp()", uses the parent axis hint as it's own.
- New: In State::slackenTopology(), in the global FSM, adds a special
operation when reaching MaximumSlack: forceOverLocals(), try to insert
the global on track containing only local segments. Should tend to
concentrate locals on a small set of shared tracks. Most useful on the
highest layers.
- New: In State::slackenTopology(), in the "MoveUp" state, try to find the
more appropriate segment to move up (Manipulator::desaturate()).
Effectively move up the longest segment fully enclosing the one we are
processing.
- New: In State::slackenTopology(), add a check for fully blocked segments
in the local segment FSM. Calls State::solveFullBlocked().
- New: In KiteEngine::createGlobalGraph(), decrease the vertical capacity
of one track inside the core. Helps smooth the vertical density.
- Change: In Manipulator::insertInTrack(), when a track is freed for a
to be inserted changes the priorities so that the segment is immediatly
inserted. Parallels ripeds and theirs perpandiculars are replaced
only *after*. This is the opposite of the previous behavior.
- Change: In NegociateWindow::NegociateOverlapCost(), account the costs
of terminals only for deep depth layers (M1, M2 & M3).
- Change: In RoutingEvent::insertInTrack(), expand the excluded interval
by a half-pitch (2.5l) instead of one lambda.
- Change: In State::State(), do not uses DiscardGlobal if the ripup count
exceed 5. Case of the "Strap" segments that can be ripped a lot
before changing state.
- Change: In State::_processNegociate(), no longer lock into position
(fixed) the local terminal segments as a last resort.
- Change: In RoutingEvent::_processNegociate(), no longer ripup perpandiculars
when a segment is inserted in a free space. Reduce the number of
events whithout degrading the routing quality.
- Change: In State::conflictSolve1_v1b(), if getLongestConflict() is nul,
ignore the track, the conflict must occurs on another track.
- Change: In TrackCost, add a flag support. First uses, a flags to prevent
a local of the topmost layer to ripup a global which is in moveUp
state.
- Bug: In State::solveFullBlockage(), after have been freed, reset the
segment state to "moveUp".
- Bug: In manipulator::minimize(), the axisHint was miscalculated if the
punctual span was empty.
- New: In RoutingEvent, while routing a full chip, transient full blockages
may happens due to the initial position of some perpandiculars on theirs
optimal positions. Check for it and perform a "ripupPerpandicular" with
the perpandiculars re-routed first (on normal operation, this is the
reverse).
- Change: In NegociateWindow::NegociateOverlapCost(), do not account terminals
if the layer is *above* METAL3 as it is unlikely that it's truly directly
connected to a terminal (true at least for our designs).
- Change: In RoutingEvent::State, add a "fullBlocked" flag to perform the
full blockage direction while building the state object.
- Change: In RoutingEvent::cacheAxisHint(), when the TrackSegment has parent,
uses the parent current axis position and not it's axis hint.
- Change: In Manipulator::insertInTrack(), when the overlapped segment is a
Local and is completly enclosed (shrinkLeft & shrinkRight), no longer
rip it up but force a shrink left/right instead.
- Bug: In RoutingEvent, the event sorting function was sorting in the *wrong*
order! The less prioritary first! With the correct sort, we won an
additionnal 30% in speed (total: 69%). From the "reference" time we have
a 3.2 speed-up. And we can successfully process denser designs...
- New: In BuildPowerRails, special processing for the power ring segments.
The "diagonal" of vias at each corner is causing a misbehavior of the
routing algorithm (due to fully saturated GCells in one direction).
As a temporary fix, extend the segments so they form a "square corner".
(problem arise on "d_in_i(22)").
- New: In RoutingEvent::_processNegociate, disable the "isForcedToHint()"
feature. No noticeable loss of quality or speed.
- New: In TrackElement/TrackSegment, wraps the AutoSegment parent's mechanism.
Allows to gets the DataNegociate of either the segment or it's parent.
- New: State::solveFullBlockages(), dedicated method to solves the case when
all the allowed tracks of a segment are blocked, tries to moves up
local segments and to break-up global ones.
- New: RoutingEventLoop, a more sophisticated way to detect looping.
Maintain a dynamic histogram of the last N (default 10) segments routeds,
with the count of how many times they have occurred. If that count
exeed 40, we *may* be facing a loop.
- Change: In State::conflictSolve1, implement new policy. The global segments
no more can be broken by local ones. The idea behind is that breaking
a global on the request of a local will only produce more cluttering
in the GCell. Globals must be keep straigth pass through, especially
inside near-saturated GCells. Globals breaking however can occurs at
another global's request.
- Change: In TrackCost, implement the new policy about locals segments that
cannot break globals segments. The sorting class now accept flags to
modulate the sorting function. Two options avalaibles: IgnoreAxisWeigth
(to uses for strap segments) and DiscardGlobals (to uses with locals).
- Change: In TrackCost, the "distance to fixed" have now an upper bound of
50 lambdas (no need to be greater because it means it's outside the
begin & en GCells). Take account not only of fixed segment, but also
of placed segments which makes bound.
- Bug: In Track::_check(), while calling each individual TrackSegment check,
uses it as the *first* argument of the "or", otherwise it may not be
called.
- Bug: In ProtectRoutingPad, loop over segment Collections while modificating
it was producing non-deterministic results. The fact that a collection
must be not modificated while beeing iterated is becoming a more and more
painful problem.
- Change: Big cleanup. Supress all the GCell level: GCell, GCellGrid,
GCellRoutingSet. Blockages now managed inside the BuildPower.
- New: In NegociateWindow, Histogram of GCell densities in Gnuplot
format.
- Change: Propagate renaming "obstacle" -> "blockage".
- Bug/Change: In Configuration, the value of the extensionCap was too big
(1.5 lambda), reduce to 0.5 lambda. This is a problem, the extension
should be coupled to the layer as it is not the same for each METAL.
- Bug: When using TrackElement, always uses the virtual "->isFixed()" method
instead of trying to access to "->base()->isFixed()" as the base may be
NULL in case of blockage/fixed segment.
- Change: Merge PowerRails & Blockage trans-hierarchical construction (into
PowerRails). All blockages are groupeds under "blockagenet".
Allows to remove TrackBlockage & BuildBlockages.
- Change: In KiteEngine::annotateGloblalGraph(), when routing a full chip,
ring power segments around the core must completly saturate the edges
in their segment direction. This is to prevent the global router to
use paths under the power/ground ring (may generate unsolvable configs).
- Change: In KiteEngine::annotateGloblalGraph(), when routing a full chip,
distinguish three areas: the core (65%), the corona (90%) and the pads
(100%). Capacities on the edges are sets accordingly.
- Change: In RoutingEvent, introduce an alternative algorithm for
conflictSolve1, FindPath which try to deduce the breakpoints from a
truly explorated path. Unfortunatly this gives worst results than the
Cs1Candidates method. The why should be investigated as it's a critical
point in the algorithm.
- Change: In Manipulator::ripupPerpandicular(), when a caged perpandicular
is encountered, instead of just "stopping", rip it up and change is
axis hint (actually increase) it's axis hint so it stands a chance to
go outside the track with an obstacle.
- Change: In RoutingEvent/State::slackenTopology(), allow move up of local
segments when they are tightly constrained *and* blocked (cageds).
Partial modification of functions calls from booleans to flags.
- Bug: In NegociateWindow::NegociateOverlapCost, check for fixed segments
before trying to get DataNegociate. The lack of DataNegociate cause the
TrackElement to be discarted. It's a failsafe behavior, but it leads to
overlaps.
- Bug: In ProtectRoutingPad, in Pad Cells only, *do not* protect RoutingPad
to avoid the edge capacity over the pad to decrease to zero. This is
due to unused RoutingPads being accounted as blockages.
- New: In Configuration & RoutingEvent, adds support for minimal break length
by layer (METAL1 through METAL7).
- New: Added measures for final detailed wirelength, wirelength expansion ratio,
failed wirelength, number of processeds events and number of unique events.
- Change: In TrackCost, the ripup cost is now computed from a mix between
the number of ripup *and* the data state. Restore a correct balance
of the ripping up between segments.
- New: In RoutingEvent::State::conflictSolve1(), when a very long Segment
enters in conflictSolve1(), still prefers to move it up in one piece,
but if it's not possible, it is unlikely that it will overlap with
only one or two other Segment. In that cases, blindy break it up in
the middle.
- New: In RoutingEvent::Manipulator, another overload of makeDogLeg to
allow break up at one precise point.
- Change/Bug: In RoutingEvent::Manipulator::Relax(), reset the state of
the non-overlapping Segment fragments instead of inheriting blindly
of the state of the breaked up one. Gives some supplemental slack
that allows to converge.
- Note: Now the strategy is to disable the uses of GCellRoutingSet by
setting the expand step to 0.99. This induces only a slight slow
dow and memory increase and passes all the tests...
- Mark: This revision passes all the benches except for idct, vld &
ieee_division.
- Bug: In top CMakeLists.txt the SETUP_PROJECT_PATHS was not inserting
X_USER_TOP *before* X_TOP, thus potentially allowing an obsolete
system-wide configuration to shadow an up-to-date local one.
* ./kite:
- Change: In Configuration/KiteEngine, propagate changes in Configuration
in Kite. Note that a *second* singleton is created at Kite level.
- New: KiteEngine::printConfiguration() method to allow quick checks of
the router's main parameters.
- New: ConfigurationWidget to display/sets all the revelant Katabatic/
Kite configuration parameters. To be used with the TabSettings from
the ControllerWidget.
The ConfigurationWidget is bound to the default (static) Kite
configuration, as once it has been run you obviously cannot change the
router's settings.
- Change: In TrackFixedSegment, when the source/target coordinates of the
fixed segment are outside the grid (i.e. the Cell Abutment Box) truncate
them.
- Change: In GCell::addTrackSegment, when looking for overlap between the
to be inserted TrackSegment and Blockage segment, shrink the Blockage
span by one DbU to avoid false overlap when they are just edges connexes.
- Bug: In RoutingEvent::slackenTopology, in the local segment FSM, looping
on the MaximumSlack state. Now also checks the state count.
- Change: In TrackSegment::canMoveUp() & canPivotUp() adds the "reserve"
parameter and uses it in RoutingEvent to prevent over saturated GCells.
- Change: In Configuration, parameters adjustements: expandStep increased to
0.40. Makes RoutingSets biggers but no noticeable slow down.
- Change: In TrackSegment, _postDogLeg() & _postModify(), add segments to
"Ring" if they are directly connected to a RoutingPad. To maintain the
"border effect" after a topological slackening.
- Change: In GraphicKiteEngine, in the global routing relateds menus, suppress
the mention of "solution" as it might be unclear for an average user.
(just "Save Global Routing" or "Load Global Routing").
- Change: In RoutingEvent, conflictSolve1() tries to move up "other" blocking
segment before trying to break "self".
- Bug: In TrackSegment::reschedule(), remove from Track segments that are
now in a greater order than the current one. This bug was shown in the
TrackSegment::_postModify() method after a desalignate.
It was causing overlaps and/or routing completion false reports.
- Cleanup: In GCell, remove the DyKeyQueue support as it's been moved
to Katabatic level.
- Change: In NegociateWidow, increase the event stop thresold to 4M events
to be able to process the eFPGA matrix 16x16 (150K gates).
./coriolis/src/crlcore,
./coriolis/src/knik,
./coriolis/src/katabatic,
./coriolis/src/kite,
./coriolis/src/equinox,
./coriolis/src/solstice,
./coriolis/src/ispd:
- SVN MOVE: Source tree simplification & uniformisation. Now all tools
are at the same level, directly under the root of the repository.
No more "coriolis/src".
Jean-Paul Chaput