* Change: In Hurricane Commons.h, even when cdebug print nothing, it slow
down the program (three times for Kite!). Create a macro cdebug_log
which calls cdebug *only* if the debug level is active.
* Change: In Hurricane, in Flags add operator overload for "int" type
and not only "unsigned int". Otherwise the compiler complaints about
ambiguous overload when using enum values which are considered as
"int".
Simpler code for the BaseFlags::contains() method.
Added implicit conversion from BaseFlags toward bool type.
* Change: In Hurricane, in Commons, complete replacement of the previous
two trace systems (trace & ltrace) by a stream-based one.
As it is a true object it is much less fragile than the one based
on defines (but maybe a little slower).
Define a reservation table for the trace levels for all the
Coriolis & Chams components.
* Change: All tools, use the new trace system.
* Bug: In Knik, in Vertex the _flags attribute was unitialized.
This flag is, in particular, telling if the vertex is blocked.
That it was taking a random value, was leading to errors and
unreproductible behaviors. I hope this will make the global
router deterministic at last. This problem did show in the
Stratus FIFO generator.
* Bug: In Knik, in Vertex, remove the unused _boundingBox attribute.
* Change: In vlsispad, in Dots, add an enable/disable flag because when printing
into a cmess, it is only the base class ostream which is took into account
as none of it's methods are virtuals (silly me).
* Bug: In Etesian, print into cmess instead of cout and make use of the Dots
enabling feature.
* New: In Katabatic, added AutoContact::migrateConstraintBox() to transfert
constraint from one contact to another.
New flag for AutoContact, CntWeakTerminal for AutoContact which are at
the other of a segment directly connected to a terminal. They may hold
special constraints in case of caged terminals (computed in Kite).
In AutoHorizontal & AutoVertical, in ::_makeDogleg(), transfert flags
and constraints when breaking a segment directly connected to a terminal.
* New: In Kite, in protectCagedTerminals(), uses cross constraint on the
AutoContact opposite to the ContactTerminal (CntWeakTerminal) instead
of moving up one terminal over two consecutives. This is simpler without
degrading the routing quality.
* Change: In Knik, in Vertex, add a "blocked" flag to signal disabled
vertexes in the grid (must not be used by the global router).
Modificate the Graph::getVertex() method so that when a vertex
is geometrically queried, if is a blocked one, return a non-blocked
neighbor. This mechanism is introduced to, at last, prevent the
global router to go *under* the pad in case of a commplete chip.
* New: In Katabatic, in AutoSegment, a new state has been added: "reduced".
A reduced segment is in the same layer as it's perpandiculars.
To be reduced, a segments has to be connected on source & target to
AutoContactTurn, both of the perpandiculars must be of the same layer
(below or above) and it's length must not exceed one pitch in the
perpandicular direction.
To reduce an AutoSegment, call ::reduce() and to revert the state,
call ::raise(). Two associated predicates are associated:
::canReduce() and ::mustRaise().
Note: No two adjacent segments can be reduced at the same time.
* Bug: In Katabatic, in GCellTopology, add a new method ::doRp_AccessPad()
to connect to the pads. Create wiring, fixed and non managed by
Katabatic, to connect the pad connector layer to the lowest routing
layers (depth 1 & 2). The former implementation was sometimes leading
to gaps (sheared contact) that *must not* occurs during the building
stage.
Remark: This bug did put under the light the fact that the initial
wiring must be created without gaps. Gaps are closed by making doglegs
on contacts. But this mechanism could only work when the database if
fully initialised (the cache is up to date). Otherwise various problems
arise, in the canonization process for example.
* New: In Katabatic, in AutoContactTerminal::getNativeConstraintBox(),
when anchored on a RoutingPad, now take account the potential rotation
of the Path's transformation. Here again, for the chip's pads.
* New: In Kite, support for reduced AutoSegment. TrackSegment associateds
to reduced AutoSegment are *not* inserted into track to become
effectively invisibles. When a segment becomes reduced, a TrackEvent
is generated to remove it. Conversely when it is raised a RoutingEvent
is created/rescheduled to insert it. All this is mostly managed inside
the Session::revalidate() method.
* New: In Kite, in KiteEngine::createGlobalGraph(), in case of a chip,
mark all global routing vertexes (Knik) that are under a pad, as blockeds.
* Bug: In Cumulus, in PadsCorona.Side.getAxis(), inversion between X and
Y coordinate of the chip size. Did not show until a non-square chip
was routed (i.e. our MIPS R3000).
* Change: In Stratus1, in st_placement.py add the ClockBuffer class for
backward compatibility with the MIPS32 bench. Have to review this
functionnality coming from the deprecated placeAndroute.py.
In st_instance.py, no longer creates the Plug ring of a Net.
In my opinion it just clutter the display until the P&R is called.
Can re-enable later as an option (in Unicorn).
* Change: In Unicorn, in cgt.py, more reliable way of loading then running
user supplied scripts. Borrowed from alliance-checker-toolkit doChip.py .
* Bug: In Knik, in Graph::Dijsktra(), when all the RoutingPads of a net
are aligned, the seaching area is only one vertex wide (in width or
height). That is, it contains only one edge in it's main direction.
So when that sub-graph contains a fully blocked edge, Dijkstra cannot
find a path... Now enlarge the one edge graph to at least three by
inflating the search area. Curiously a code for similar purpose seems
to be present, but commented.
* New: In Hurricane, in DebugSession, add a new method to activate the
trace inconditionally with a certain level.
* New: In Hurricane, in HyperNet, allow copy construction as there is
no reason to disallow it and we need it now.
* New: In Hurricane, in Cell::flattenNets(), add a new option to prevent
the flattening of the clock net(s). For more safety perform the
DeepNet creation *outside* the Collection loop.
* Bug: In Hurricane, in Cell_HyperNetRootNetOccurrences, skip the
DeepNets because they are the result of another flattening operation.
* New: In Isobar, in PyBasicLayer, export C++ method getBlockageLayer().
* New: In Isobar, in PyRoutingGauge, export C++ method getLayerPitch(),
needed by Cumulus plugins.
* New: In Etesian, EtesianEngine::findYSpin() to look for the Y orientation
if some cells are already placed. Typically the buffers of a clock
tree.
Pass the correct orienation to row_compatible_orientation().
Do not try to add feeds in the ISPD05 benchmarks. For now the
benchmarks are detected through their names (unreliable).
* Change: In Knik, in KnikEngine::initGlobalRouting(), allow the clock
to be routed as an ordinary signal when the clock tree is not
used.
* New: In Kite, in BuildPowerRails, management & detection for the
pre-routed clock.
In KiteEngine constructor, early initialization of the blockage
net to avoid later troubles in BuildPowerRails.
* New: In Cumulus, in ChipPlugin, add support for Etesian plus new
configuration parameter 'clockTree.placerEngine' to select between
Mauka and Etesian.
* New: In Cumulus, in BlockCorona, add blockages in the vertical sides
in the vertical layer to prevent the router to use the vertical
tracks in under the prower lines (for example, blockage in M3
with power line M5).
In Cumulus, in ChipConf add attribute to access the blockage net.
* New: In Cumulus, when the clock tree is disabled, do not generate
the last rail around the block (the clock rail).
* Bug: In Cumulus, in ChipConf use the clock pad to guess the clock
signals and *not* the power pad.
Add more meaningful error messages if a pad global signal is
not found (implicit connexion by name).
* Bug: In Cumulus, in ClockTree, compute correctly the cells Y spin,
that is *from the bottom of the cell AB* (not from 'zero').
* New: In Bootstrap, added support for RHEL7 in ccb.py & coriolisEnv.py.
In coriolisEnv.py, under RHEL6, directly launch a shell under
devtoolset-2 (do not put it in .bashrc).
* Change: All tools, compliance with Clang 3.2.
- Call using namespace std *after* at least one include of std has
been used.
- In Utilities.h define both const and non-cont overload for operator<<
of mstream (this was the true cause of the Banner display problem).
- No longer use defaults arguments in templates, instead create two
separated overloads.
- Put template static attributes allocation outside of namespaces
(this one I'm not sure how to justify).
- Protect by NDEBUG variables that are only needed in assert().
- In PyInterval getUnion() & getIntersection() we where silently
overwriting the "self" object (interval).
- In Mauka, *do no* overload _postCreate() and add an argument, breaks
the virtual function mechanism. Rename it into _maukaPostCreate().
Idem for SubRow().
- In Katabatic::GCell(), invalidate() overload a Go funtion but is
used for different purpose at this level. Rename it invalidateCt()
(invalidate *ConTents*).
- Miscellaneous small cleanup.
* Change: In bootstrap, in ccb, builder and coriolisEnv.py, correct detection
of the windows architecture (32 or 64 bits). Under Cygwin, add the
directory of the dll into the PATH also. Uses "site-package" as the
location of Python modules (*not* "dist-package").
Use gnu++0x under Cygwin instead of c++11.
* Change: In Etesian, if Coloquinte is not found, do not stop the compilation,
just disable the tool altogether.
* Change: In Hurricane, In Backtrace, disable backtrace under Cygwin as it
uses features specific of the glibc.
* Change: In Knik, use HUGE_VAL instead of HUGE (not present under Cygwin),
add the <cmath> header.
* Change: In Unicorn, Coloquinte no longer stops the compilation.
* Change: In Kite, Katabatic & Knik, express the number of tracks truly
available to the global router by the number of tracks reserved to
the local routage inside a GCell. Replace the ratio parameter
"hEdgeCapacity" by an integer parameter "hTracksReservedLocal"
(duplicate for verticals).
It is more explicit to give directly the number of tracks that
are to be used locally, and potentially saves us from rouding
problems when calculating the number of availables tracks.
Note: we cannot do that for the layer saturateRatio as it
uses the density ratio that take account local wires, leading
to fractional results.
* Change: In Katabatic, in <GCellGrid>, rename checkEdgeSaturation()
into checkEdgeOverflow(), more explicit.
* Change: In Knik, in <Graph>, display the computed capacities of the
lower left node edges (should be the same througout all the grid).
* Change: In Unicorn, in <cgt.py>, uses the new parameters names for
edge density.
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.
Details:
* New: in <bootstrap>: add support for devtoolset-2 in ccb. Run the
cmake commands through 'scl', set shell environment variables
BOOST_INCLUDEDIR & BOOST_LIBRARYDIR and disable the default
system path search.
* Change: In various flex scanners add the %nounput to suppress
compiler warnings.
* Change: Little cleanup for g++ 4.8.1 as it's more strict.
* Change: In various top CMakeLists.txt, suppress extraneous '/'
after DESTDIR.
- Bug: In Edge::getCost(), when the edge capacity is null, return maximum
cost (HUGE).
- Bug: In Knik::saveSolution(), saves only Segments that belongs to global
routing, that is in GMetalH/GMetalV.
- New: KnikEngine::computeSymbolicWireLength(), which adds to the measurments
1. - The sum of global wire length (in lambdas).
2. - The abutment box area of the design (in square lambdas).
3. - The saturation ratio of the design: wire length divided by area and
normalized by a "fully routed GCell".
- Bug: In MatrixVertex::getLineIndex() and getColumnIndex(), in regular
mode, when the Y/X coordinate is exactly on the upper right angle,
return the inferior index (instead of superior which is out of bounds).
- Change: In MatrixVertex, RoutingGrid & KnikEngine, replace the _lowerLeftX &
_lowerLeftY by a complete copy of the Cell's abutment box. Makes
easier for getLineIndex() to find out of bound coordinates.
- Change: In KnikEngine & LoadSolution add a time/memory measurement,
for Kite statistics.
./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
Gabriel Gouvine