coriolis/katana/src/NegociateWindow.cpp

659 lines
21 KiB
C++
Raw Normal View History

Anabatic transient commit 18. Port of Kite (Katana), Yeah, Baby! Yeah! * Bug: In Hurricane, in StaticObservable::getObserver(), if the slot pointer is NULL, do not try to access the owner. Returns NULL, so the caller can be aware of the situation... * Change: In Hurricane, in BreakpointWidget & ExceptionWidget some cosmetic changes (fonts and window sizes). * Bug: In Anabatic, In AutoHorizontal::getConstraints(), take into account the constraints from the source AutoContact, as it holds the constraints transmitted by the RoutingPads and sets up by propageConstraintsFromRp(). It is likely to be a bug affecting the original Katabatic as well. * Change: In Anabatic, in RawGCellsUnder(), check that the segment is not completly oustside the cell abutment box and truncate the coordinates to the part that is inside. Use the "shrink" if we reach the east/north border. * Change: In Anabatic, in Configuration, no more decorator because we will use a true derived relationship. Katana *derives* from *Anabatic* and do not *decorate* it, so the Configuration can do the same. It also implies that we directly create a Katana engine, not an Anabatic one. * Change: In Anabatic, in Session, do not allow the opening of the Session in a standalone fashion (with a static method). Instead it must be opened using the relevant method of the Anabatic/Katana engine. This ensure we are opening the right Session type. * Change: In Anabatic, in AutoSegment_Aligneds() collection the seed segment is not part of the collection by default, but will be included if the Flags::WithSelf is set. * Change: In Configuration, all the flags value are now defined in two steps. Declared in the header and initialized in the module. This is to prevent the fact that on some cases, in relation with the Python "extern C" part modules, we need a true allocated variable. It was causing weird linking problems. A side effect is that they can no longer be used as entry is switches, have to replace them by if/else. * New: In Anabatic, new GCell::getNeighborAt() utility function. * Bug: In Anabatic, in GCell::doGrid(), tag all the GCells of the grid with the grid type... Back annote all the edges capacity (north & east) with the reserved local capacity. * New: Complete portage of Kite over Anabatic. The new engine is christened "Katana" for Kite-Analogic. When it's capabilities and performances will be on a part with Kite, it is to completly replace it (and take back the "Kite" name). Preliminary tests seems to show that, contrary to intuition (because built on a more complex/slower grid), it is even slightly faster than Kite 8-).
2016-08-15 09:30:13 -05:00
// -*- C++ -*-
//
// This file is part of the Coriolis Software.
// Copyright (c) UPMC 2008-2016, All Rights Reserved
//
// +-----------------------------------------------------------------+
// | C O R I O L I S |
// | K i t e - D e t a i l e d R o u t e r |
// | |
// | Author : Jean-Paul CHAPUT |
// | E-mail : Jean-Paul.Chaput@asim.lip6.fr |
// | =============================================================== |
// | C++ Module : "./NegociateWindow.cpp" |
// +-----------------------------------------------------------------+
#include <vector>
#include <algorithm>
#include <fstream>
Anabatic transient commit 18. Port of Kite (Katana), Yeah, Baby! Yeah! * Bug: In Hurricane, in StaticObservable::getObserver(), if the slot pointer is NULL, do not try to access the owner. Returns NULL, so the caller can be aware of the situation... * Change: In Hurricane, in BreakpointWidget & ExceptionWidget some cosmetic changes (fonts and window sizes). * Bug: In Anabatic, In AutoHorizontal::getConstraints(), take into account the constraints from the source AutoContact, as it holds the constraints transmitted by the RoutingPads and sets up by propageConstraintsFromRp(). It is likely to be a bug affecting the original Katabatic as well. * Change: In Anabatic, in RawGCellsUnder(), check that the segment is not completly oustside the cell abutment box and truncate the coordinates to the part that is inside. Use the "shrink" if we reach the east/north border. * Change: In Anabatic, in Configuration, no more decorator because we will use a true derived relationship. Katana *derives* from *Anabatic* and do not *decorate* it, so the Configuration can do the same. It also implies that we directly create a Katana engine, not an Anabatic one. * Change: In Anabatic, in Session, do not allow the opening of the Session in a standalone fashion (with a static method). Instead it must be opened using the relevant method of the Anabatic/Katana engine. This ensure we are opening the right Session type. * Change: In Anabatic, in AutoSegment_Aligneds() collection the seed segment is not part of the collection by default, but will be included if the Flags::WithSelf is set. * Change: In Configuration, all the flags value are now defined in two steps. Declared in the header and initialized in the module. This is to prevent the fact that on some cases, in relation with the Python "extern C" part modules, we need a true allocated variable. It was causing weird linking problems. A side effect is that they can no longer be used as entry is switches, have to replace them by if/else. * New: In Anabatic, new GCell::getNeighborAt() utility function. * Bug: In Anabatic, in GCell::doGrid(), tag all the GCells of the grid with the grid type... Back annote all the edges capacity (north & east) with the reserved local capacity. * New: Complete portage of Kite over Anabatic. The new engine is christened "Katana" for Kite-Analogic. When it's capabilities and performances will be on a part with Kite, it is to completly replace it (and take back the "Kite" name). Preliminary tests seems to show that, contrary to intuition (because built on a more complex/slower grid), it is even slightly faster than Kite 8-).
2016-08-15 09:30:13 -05:00
#include <iomanip>
#include "hurricane/Warning.h"
#include "hurricane/Bug.h"
#include "hurricane/RoutingPad.h"
#include "hurricane/Net.h"
#include "hurricane/Cell.h"
#include "crlcore/Utilities.h"
#include "crlcore/AllianceFramework.h"
#include "crlcore/Measures.h"
#include "crlcore/Histogram.h"
#include "anabatic/AutoContact.h"
#include "katana/DataNegociate.h"
#include "katana/TrackElement.h"
#include "katana/TrackMarker.h"
#include "katana/TrackCost.h"
#include "katana/Track.h"
#include "katana/TrackSegment.h"
#include "katana/RoutingPlane.h"
#include "katana/RoutingEventQueue.h"
#include "katana/RoutingEventHistory.h"
#include "katana/RoutingEventLoop.h"
#include "katana/NegociateWindow.h"
#include "katana/KatanaEngine.h"
namespace {
using namespace std;
using namespace Hurricane;
using namespace CRL;
using namespace Katana;
void NegociateOverlapCost ( const TrackElement* segment, TrackCost& cost )
{
cdebug_log(9000,0) << "Deter| NegociateOverlapCost() " << segment << endl;
Interval intersect = segment->getCanonicalInterval();
if (not intersect.intersect(cost.getInterval())) return;
if (segment->isBlockage() or segment->isFixed()) {
cdebug_log(159,0) << "Infinite cost from: " << segment << endl;
cost.setInfinite ();
cost.setOverlap ();
cost.setHardOverlap();
cost.setBlockage ();
return;
}
if (cost.getInterval().getVMax() > intersect.getVMax()) cost.setLeftOverlap();
if (cost.getInterval().getVMin() < intersect.getVMin()) cost.setRightOverlap();
if (not intersect.contains(cost.getInterval()))
intersect.intersection( cost.getInterval() );
else {
cost.setLonguestOverlap( intersect.getSize() );
cost.setGlobalEnclosed();
}
DataNegociate* data = segment->getDataNegociate();
if (not data) return;
cost.mergeRipupCount( data->getRipupCount() );
if ( segment->isLocal() ) {
cost.mergeDataState( data->getState() );
if (data->getState() >= DataNegociate::LocalVsGlobal) {
cdebug_log(159,0) << "MaximumSlack/LocalVsGlobal for " << segment << endl;
}
}
if (segment->isGlobal()) {
cost.setOverlapGlobal();
if ( (cost.getFlags() & TrackCost::LocalAndTopDepth)
and (data->getState() >= DataNegociate::MoveUp) ) {
cost.setInfinite ();
cost.setOverlap ();
cost.setHardOverlap();
return;
}
}
cost.setOverlap();
if ( segment->isLocal()
or (cost.isForGlobal() and (Session::getRoutingGauge()->getLayerDepth(segment->getLayer()) < 3)) ) {
cdebug_log(9000,0) << "Deter| incTerminals() " << boolalpha << cost.isForGlobal() << " " << (data->getTerminals()*100) << endl;
cost.incTerminals( data->getTerminals()*100 );
} else {
cdebug_log(9000,0) << "Deter| isForGlobal() " << boolalpha << cost.isForGlobal() << endl;
}
cdebug_log(159,0) << "| Increment Delta: " << DbU::getValueString(intersect.getSize()) << endl;
cost.incDelta( intersect.getSize() );
}
void loadRoutingPads ( NegociateWindow* nw )
{
AllianceFramework* af = AllianceFramework::get ();
RoutingGauge* rg = nw->getKatanaEngine()->getConfiguration()->getRoutingGauge();
for( Net* net : nw->getCell()->getNets() ) {
if (net->getType() == Net::Type::POWER ) continue;
if (net->getType() == Net::Type::GROUND) continue;
if (net->getType() == Net::Type::CLOCK ) continue;
if (af->isBLOCKAGE(net->getName())) continue;
for( RoutingPad* rp : net->getRoutingPads() ) {
size_t depth = rg->getLayerDepth(rp->getLayer());
if (depth > 0) continue;
if (depth == 0)
TrackMarker::create( rp, 1 );
}
}
}
} // Anonymous namespace.
namespace Katana {
using std::ofstream;
Anabatic transient commit 18. Port of Kite (Katana), Yeah, Baby! Yeah! * Bug: In Hurricane, in StaticObservable::getObserver(), if the slot pointer is NULL, do not try to access the owner. Returns NULL, so the caller can be aware of the situation... * Change: In Hurricane, in BreakpointWidget & ExceptionWidget some cosmetic changes (fonts and window sizes). * Bug: In Anabatic, In AutoHorizontal::getConstraints(), take into account the constraints from the source AutoContact, as it holds the constraints transmitted by the RoutingPads and sets up by propageConstraintsFromRp(). It is likely to be a bug affecting the original Katabatic as well. * Change: In Anabatic, in RawGCellsUnder(), check that the segment is not completly oustside the cell abutment box and truncate the coordinates to the part that is inside. Use the "shrink" if we reach the east/north border. * Change: In Anabatic, in Configuration, no more decorator because we will use a true derived relationship. Katana *derives* from *Anabatic* and do not *decorate* it, so the Configuration can do the same. It also implies that we directly create a Katana engine, not an Anabatic one. * Change: In Anabatic, in Session, do not allow the opening of the Session in a standalone fashion (with a static method). Instead it must be opened using the relevant method of the Anabatic/Katana engine. This ensure we are opening the right Session type. * Change: In Anabatic, in AutoSegment_Aligneds() collection the seed segment is not part of the collection by default, but will be included if the Flags::WithSelf is set. * Change: In Configuration, all the flags value are now defined in two steps. Declared in the header and initialized in the module. This is to prevent the fact that on some cases, in relation with the Python "extern C" part modules, we need a true allocated variable. It was causing weird linking problems. A side effect is that they can no longer be used as entry is switches, have to replace them by if/else. * New: In Anabatic, new GCell::getNeighborAt() utility function. * Bug: In Anabatic, in GCell::doGrid(), tag all the GCells of the grid with the grid type... Back annote all the edges capacity (north & east) with the reserved local capacity. * New: Complete portage of Kite over Anabatic. The new engine is christened "Katana" for Kite-Analogic. When it's capabilities and performances will be on a part with Kite, it is to completly replace it (and take back the "Kite" name). Preliminary tests seems to show that, contrary to intuition (because built on a more complex/slower grid), it is even slightly faster than Kite 8-).
2016-08-15 09:30:13 -05:00
using std::cerr;
using std::endl;
using std::setw;
using std::left;
using std::right;
using std::setprecision;
using Hurricane::Warning;
using Hurricane::Bug;
using Hurricane::tab;
using Hurricane::ForEachIterator;
using CRL::Histogram;
using CRL::addMeasure;
using Anabatic::AutoContact;
using Anabatic::AutoSegmentLut;
using Anabatic::perpandicularTo;
// -------------------------------------------------------------------
// Class : "NegociateWindow".
NegociateWindow::NegociateWindow ( KatanaEngine* katana )
: _flags (Flags::NoFlags)
, _interrupt (false)
, _katana (katana)
, _gcells ()
, _segments ()
, _eventQueue ()
, _eventHistory()
, _eventLoop (10,50)
{ }
NegociateWindow* NegociateWindow::create ( KatanaEngine* katana )
{
NegociateWindow* negociateWindow = new NegociateWindow ( katana );
return negociateWindow;
}
NegociateWindow::~NegociateWindow ()
{ }
void NegociateWindow::destroy ()
{ delete this; }
Cell* NegociateWindow::getCell () const
{ return _katana->getCell(); }
void NegociateWindow::setGCells ( const vector<GCell*>& gcells )
{
_gcells = gcells;
loadRoutingPads( this );
Session::revalidate();
for ( auto element : Session::getKatanaEngine()->_getAutoSegmentLut() ) {
TrackElement* segment = Session::lookup( element.second );
if (segment) segment->getDataNegociate()->update();
}
_statistics.setGCellsCount( _gcells.size() );
}
void NegociateWindow::addRoutingEvent ( TrackElement* segment, uint32_t level )
Anabatic transient commit 18. Port of Kite (Katana), Yeah, Baby! Yeah! * Bug: In Hurricane, in StaticObservable::getObserver(), if the slot pointer is NULL, do not try to access the owner. Returns NULL, so the caller can be aware of the situation... * Change: In Hurricane, in BreakpointWidget & ExceptionWidget some cosmetic changes (fonts and window sizes). * Bug: In Anabatic, In AutoHorizontal::getConstraints(), take into account the constraints from the source AutoContact, as it holds the constraints transmitted by the RoutingPads and sets up by propageConstraintsFromRp(). It is likely to be a bug affecting the original Katabatic as well. * Change: In Anabatic, in RawGCellsUnder(), check that the segment is not completly oustside the cell abutment box and truncate the coordinates to the part that is inside. Use the "shrink" if we reach the east/north border. * Change: In Anabatic, in Configuration, no more decorator because we will use a true derived relationship. Katana *derives* from *Anabatic* and do not *decorate* it, so the Configuration can do the same. It also implies that we directly create a Katana engine, not an Anabatic one. * Change: In Anabatic, in Session, do not allow the opening of the Session in a standalone fashion (with a static method). Instead it must be opened using the relevant method of the Anabatic/Katana engine. This ensure we are opening the right Session type. * Change: In Anabatic, in AutoSegment_Aligneds() collection the seed segment is not part of the collection by default, but will be included if the Flags::WithSelf is set. * Change: In Configuration, all the flags value are now defined in two steps. Declared in the header and initialized in the module. This is to prevent the fact that on some cases, in relation with the Python "extern C" part modules, we need a true allocated variable. It was causing weird linking problems. A side effect is that they can no longer be used as entry is switches, have to replace them by if/else. * New: In Anabatic, new GCell::getNeighborAt() utility function. * Bug: In Anabatic, in GCell::doGrid(), tag all the GCells of the grid with the grid type... Back annote all the edges capacity (north & east) with the reserved local capacity. * New: Complete portage of Kite over Anabatic. The new engine is christened "Katana" for Kite-Analogic. When it's capabilities and performances will be on a part with Kite, it is to completly replace it (and take back the "Kite" name). Preliminary tests seems to show that, contrary to intuition (because built on a more complex/slower grid), it is even slightly faster than Kite 8-).
2016-08-15 09:30:13 -05:00
{
DataNegociate* data = segment->getDataNegociate();
if (not data or not data->hasRoutingEvent())
_eventQueue.add( segment, level );
else
cerr << Bug( "NegociateWidow::addRoutingEvent(): Try to adds twice the same TrackElement event."
"\n %p:%s."
, (void*)segment->base()->base()
, getString(segment).c_str()
) << endl;
}
TrackElement* NegociateWindow::createTrackSegment ( AutoSegment* autoSegment, Flags flags )
Anabatic transient commit 18. Port of Kite (Katana), Yeah, Baby! Yeah! * Bug: In Hurricane, in StaticObservable::getObserver(), if the slot pointer is NULL, do not try to access the owner. Returns NULL, so the caller can be aware of the situation... * Change: In Hurricane, in BreakpointWidget & ExceptionWidget some cosmetic changes (fonts and window sizes). * Bug: In Anabatic, In AutoHorizontal::getConstraints(), take into account the constraints from the source AutoContact, as it holds the constraints transmitted by the RoutingPads and sets up by propageConstraintsFromRp(). It is likely to be a bug affecting the original Katabatic as well. * Change: In Anabatic, in RawGCellsUnder(), check that the segment is not completly oustside the cell abutment box and truncate the coordinates to the part that is inside. Use the "shrink" if we reach the east/north border. * Change: In Anabatic, in Configuration, no more decorator because we will use a true derived relationship. Katana *derives* from *Anabatic* and do not *decorate* it, so the Configuration can do the same. It also implies that we directly create a Katana engine, not an Anabatic one. * Change: In Anabatic, in Session, do not allow the opening of the Session in a standalone fashion (with a static method). Instead it must be opened using the relevant method of the Anabatic/Katana engine. This ensure we are opening the right Session type. * Change: In Anabatic, in AutoSegment_Aligneds() collection the seed segment is not part of the collection by default, but will be included if the Flags::WithSelf is set. * Change: In Configuration, all the flags value are now defined in two steps. Declared in the header and initialized in the module. This is to prevent the fact that on some cases, in relation with the Python "extern C" part modules, we need a true allocated variable. It was causing weird linking problems. A side effect is that they can no longer be used as entry is switches, have to replace them by if/else. * New: In Anabatic, new GCell::getNeighborAt() utility function. * Bug: In Anabatic, in GCell::doGrid(), tag all the GCells of the grid with the grid type... Back annote all the edges capacity (north & east) with the reserved local capacity. * New: Complete portage of Kite over Anabatic. The new engine is christened "Katana" for Kite-Analogic. When it's capabilities and performances will be on a part with Kite, it is to completly replace it (and take back the "Kite" name). Preliminary tests seems to show that, contrary to intuition (because built on a more complex/slower grid), it is even slightly faster than Kite 8-).
2016-08-15 09:30:13 -05:00
{
cdebug_log(159,1) << "NegociateWindow::createTrackSegment() - " << autoSegment << endl;
// Special case: fixed AutoSegments must not interfere with blockages.
// Ugly: uses of getExtensionCap().
if (autoSegment->isFixed()) {
RoutingPlane* plane = Session::getKatanaEngine()->getRoutingPlaneByLayer(autoSegment->getLayer());
Track* track = plane->getTrackByPosition( autoSegment->getAxis() );
size_t begin;
size_t end;
Interval fixedSpan;
Interval blockageSpan;
autoSegment->getCanonical( fixedSpan );
fixedSpan.inflate( Session::getExtensionCap(autoSegment->getLayer())-1 );
track->getOverlapBounds( fixedSpan, begin, end );
for ( ; (begin < end) ; begin++ ) {
TrackElement* other = track->getSegment(begin);
cdebug_log(159,0) << "| overlap: " << other << endl;
if (not other->isBlockage()) continue;
other->getCanonical( blockageSpan );
blockageSpan.inflate( Session::getExtensionCap(autoSegment->getLayer()) );
cdebug_log(159,0) << " fixed:" << fixedSpan << " vs. blockage:" << blockageSpan << endl;
if (not fixedSpan.intersect(blockageSpan)) continue;
// Overlap between fixed & blockage.
cdebug_log(159,0) << "* Blockage overlap: " << autoSegment << endl;
Session::destroyRequest( autoSegment );
cerr << Warning( "Overlap between fixed %s and blockage at %s."
, getString(autoSegment).c_str()
, getString(blockageSpan).c_str() ) << endl;
cdebug_tabw(159,-1);
return NULL;
}
}
Interval span;
autoSegment = autoSegment->getCanonical( span );
bool created;
TrackElement* trackSegment = TrackSegment::create( autoSegment, NULL, created );
if (not (flags & Flags::LoadingStage))
cdebug_log(159,0) << "* lookup: " << autoSegment << endl;
if (created) {
cdebug_log(159,0) << "* " << trackSegment << endl;
RoutingPlane* plane = Session::getKatanaEngine()->getRoutingPlaneByLayer(autoSegment->getLayer());
Track* track = plane->getTrackByPosition ( autoSegment->getAxis() );
Interval uside = autoSegment->getAutoSource()->getGCell()->getSide( perpandicularTo(autoSegment->getDirection()) );
if (track->getAxis() > uside.getVMax()) track = track->getPreviousTrack();
if (track->getAxis() < uside.getVMin()) track = track->getNextTrack();
cdebug_log(159,0) << "* GCell U-side " << uside << endl;
cdebug_log(159,0) << "* " << plane << endl;
cdebug_log(159,0) << "* " << track << endl;
trackSegment->setAxis( track->getAxis(), Anabatic::SegAxisSet );
trackSegment->invalidate();
if (trackSegment->isFixed()) {
Session::addInsertEvent( trackSegment, track );
} else {
_segments.push_back( trackSegment );
}
}
if (not created and not (flags & Flags::LoadingStage)) {
cdebug_log(159,0) << "TrackSegment already exists (and not in loading stage)." << endl;
}
cdebug_tabw(159,-1);
return trackSegment;
}
double NegociateWindow::computeWirelength ()
{
set<TrackElement*> accounteds;
double totalWL = 0.0;
for ( size_t igcell=0 ; igcell<_gcells.size() ; ++igcell ) {
double gcellWL = 0.0;
Segment* segment;
TrackElement* trackSegment;
const vector<AutoContact*>& contacts = _gcells[igcell]->getContacts();
for ( size_t i=0 ; i<contacts.size() ; i++ ) {
for( Hook* hook : contacts[i]->getBodyHook()->getSlaveHooks() ) {
Hook* sourceHook = dynamic_cast<Segment::SourceHook*>(hook);
if (not sourceHook) continue;
segment = dynamic_cast<Segment*>(sourceHook->getComponent());
trackSegment = Session::lookup( segment );
if (trackSegment) {
if (accounteds.find(trackSegment) != accounteds.end()) continue;
accounteds.insert( trackSegment );
gcellWL += DbU::getLambda( trackSegment->getLength() );
}
}
}
// Partial sum to limit rounding errors.
totalWL += gcellWL;
}
return totalWL;
}
void NegociateWindow::_createRouting ( Anabatic::GCell* gcell )
{
cdebug_log(159,1) << "NegociateWindow::_createRouting() - " << gcell << endl;
Segment* segment;
AutoSegment* autoSegment;
cdebug_log(159,0) << "AutoSegments from AutoContacts" << endl;
const vector<AutoContact*>& contacts = gcell->getContacts();
for ( size_t i=0 ; i<contacts.size() ; i++ ) {
for( Component* component : contacts[i]->getSlaveComponents() ) {
segment = dynamic_cast<Segment*>(component);
autoSegment = Session::base()->lookup( segment );
cdebug_log(159,0) << autoSegment << endl;
if (autoSegment and autoSegment->isCanonical()) {
createTrackSegment( autoSegment, Flags::LoadingStage );
}
}
}
cdebug_log(159,0) << "_segments.size():" << _segments.size() << endl;
cdebug_tabw(159,-1);
}
void NegociateWindow::_pack ( size_t& count, bool last )
{
uint64_t limit = _katana->getEventsLimit();
uint32_t pushStage = RoutingEvent::getStage();
Anabatic transient commit 18. Port of Kite (Katana), Yeah, Baby! Yeah! * Bug: In Hurricane, in StaticObservable::getObserver(), if the slot pointer is NULL, do not try to access the owner. Returns NULL, so the caller can be aware of the situation... * Change: In Hurricane, in BreakpointWidget & ExceptionWidget some cosmetic changes (fonts and window sizes). * Bug: In Anabatic, In AutoHorizontal::getConstraints(), take into account the constraints from the source AutoContact, as it holds the constraints transmitted by the RoutingPads and sets up by propageConstraintsFromRp(). It is likely to be a bug affecting the original Katabatic as well. * Change: In Anabatic, in RawGCellsUnder(), check that the segment is not completly oustside the cell abutment box and truncate the coordinates to the part that is inside. Use the "shrink" if we reach the east/north border. * Change: In Anabatic, in Configuration, no more decorator because we will use a true derived relationship. Katana *derives* from *Anabatic* and do not *decorate* it, so the Configuration can do the same. It also implies that we directly create a Katana engine, not an Anabatic one. * Change: In Anabatic, in Session, do not allow the opening of the Session in a standalone fashion (with a static method). Instead it must be opened using the relevant method of the Anabatic/Katana engine. This ensure we are opening the right Session type. * Change: In Anabatic, in AutoSegment_Aligneds() collection the seed segment is not part of the collection by default, but will be included if the Flags::WithSelf is set. * Change: In Configuration, all the flags value are now defined in two steps. Declared in the header and initialized in the module. This is to prevent the fact that on some cases, in relation with the Python "extern C" part modules, we need a true allocated variable. It was causing weird linking problems. A side effect is that they can no longer be used as entry is switches, have to replace them by if/else. * New: In Anabatic, new GCell::getNeighborAt() utility function. * Bug: In Anabatic, in GCell::doGrid(), tag all the GCells of the grid with the grid type... Back annote all the edges capacity (north & east) with the reserved local capacity. * New: Complete portage of Kite over Anabatic. The new engine is christened "Katana" for Kite-Analogic. When it's capabilities and performances will be on a part with Kite, it is to completly replace it (and take back the "Kite" name). Preliminary tests seems to show that, contrary to intuition (because built on a more complex/slower grid), it is even slightly faster than Kite 8-).
2016-08-15 09:30:13 -05:00
RoutingEvent::setStage( RoutingEvent::Pack );
RoutingEventQueue packQueue;
//for ( size_t i = (count > 600) ? count-600 : 0
// ; (i<_eventHistory.size()-(last ? 0 : 100)) and not isInterrupted() ; i++ ) {
for ( size_t i=0 ; i<_eventHistory.size() ; ++i ) {
RoutingEvent* event = _eventHistory.getNth(i);
if ( event and not event->isCloned() ) {
cerr << "Cloned:" << event->isCloned()
<< " UTurn:" << event->getSegment()->isUTurn() << " " << event->getSegment() << endl;
}
if ( event and not event->isCloned() and event->getSegment()->isUTurn() ) {
event->reschedule( packQueue, 0 );
}
}
packQueue.commit();
while ( not packQueue.empty() and not isInterrupted() ) {
RoutingEvent* event = packQueue.pop();
if (tty::enabled()) {
cmess2 << " <pack.event:" << tty::bold << setw(8) << setfill('0')
<< RoutingEvent::getProcesseds() << tty::reset
<< " remains:" << right << setw(8) << setfill('0')
<< packQueue.size() << ">"
<< setfill(' ') << tty::reset << tty::cr;
cmess2.flush();
} else {
cmess2 << " <pack.event:" << setw(8) << setfill('0')
<< RoutingEvent::getProcesseds() << setfill(' ') << " "
<< event->getEventLevel() << ":" << event->getPriority() << "> "
<< event->getSegment()
<< endl;
cmess2.flush();
}
event->process( packQueue, _eventHistory, _eventLoop );
if (RoutingEvent::getProcesseds() >= limit) setInterrupt( true );
}
// Count will be wrong!
RoutingEvent::setStage( pushStage );
}
size_t NegociateWindow::_negociate ()
{
cdebug_log(9000,0) << "Deter| NegociateWindow::_negociate()" << endl;
cdebug_log(159,1) << "NegociateWindow::_negociate() - " << _segments.size() << endl;
cmess1 << " o Negociation Stage." << endl;
unsigned long limit = _katana->getEventsLimit();
bool profiling = _katana->profileEventCosts();
ofstream ofprofile;
if (profiling) ofprofile.open( "katana.profile.txt" );
Anabatic transient commit 18. Port of Kite (Katana), Yeah, Baby! Yeah! * Bug: In Hurricane, in StaticObservable::getObserver(), if the slot pointer is NULL, do not try to access the owner. Returns NULL, so the caller can be aware of the situation... * Change: In Hurricane, in BreakpointWidget & ExceptionWidget some cosmetic changes (fonts and window sizes). * Bug: In Anabatic, In AutoHorizontal::getConstraints(), take into account the constraints from the source AutoContact, as it holds the constraints transmitted by the RoutingPads and sets up by propageConstraintsFromRp(). It is likely to be a bug affecting the original Katabatic as well. * Change: In Anabatic, in RawGCellsUnder(), check that the segment is not completly oustside the cell abutment box and truncate the coordinates to the part that is inside. Use the "shrink" if we reach the east/north border. * Change: In Anabatic, in Configuration, no more decorator because we will use a true derived relationship. Katana *derives* from *Anabatic* and do not *decorate* it, so the Configuration can do the same. It also implies that we directly create a Katana engine, not an Anabatic one. * Change: In Anabatic, in Session, do not allow the opening of the Session in a standalone fashion (with a static method). Instead it must be opened using the relevant method of the Anabatic/Katana engine. This ensure we are opening the right Session type. * Change: In Anabatic, in AutoSegment_Aligneds() collection the seed segment is not part of the collection by default, but will be included if the Flags::WithSelf is set. * Change: In Configuration, all the flags value are now defined in two steps. Declared in the header and initialized in the module. This is to prevent the fact that on some cases, in relation with the Python "extern C" part modules, we need a true allocated variable. It was causing weird linking problems. A side effect is that they can no longer be used as entry is switches, have to replace them by if/else. * New: In Anabatic, new GCell::getNeighborAt() utility function. * Bug: In Anabatic, in GCell::doGrid(), tag all the GCells of the grid with the grid type... Back annote all the edges capacity (north & east) with the reserved local capacity. * New: Complete portage of Kite over Anabatic. The new engine is christened "Katana" for Kite-Analogic. When it's capabilities and performances will be on a part with Kite, it is to completly replace it (and take back the "Kite" name). Preliminary tests seems to show that, contrary to intuition (because built on a more complex/slower grid), it is even slightly faster than Kite 8-).
2016-08-15 09:30:13 -05:00
_eventHistory.clear();
_eventQueue.load( _segments );
cmess2 << " <queue:" << right << setw(8) << setfill('0') << _eventQueue.size() << ">" << endl;
if (cdebug.enabled(9000)) _eventQueue.dump();
size_t count = 0;
RoutingEvent::setStage( RoutingEvent::Negociate );
while ( not _eventQueue.empty() and not isInterrupted() ) {
RoutingEvent* event = _eventQueue.pop();
if (tty::enabled()) {
cmess2 << " <event:" << tty::bold << right << setw(8) << setfill('0')
<< RoutingEvent::getProcesseds() << tty::reset
<< " remains:" << right << setw(8) << setfill('0')
<< _eventQueue.size()
<< setfill(' ') << tty::reset << ">" << tty::cr;
cmess2.flush ();
} else {
cmess2 << " <event:" << right << setw(8) << setfill('0')
<< RoutingEvent::getProcesseds() << setfill(' ') << " "
<< event->getEventLevel() << ":" << event->getPriority() << "> "
<< event->getSegment()
<< endl;
cmess2.flush();
}
if (ofprofile.is_open()) {
size_t depth = _katana->getConfiguration()->getLayerDepth( event->getSegment()->getLayer() );
if (depth < 6) {
ofprofile << setw(10) << right << count << " ";
for ( size_t i=0 ; i<6 ; ++i ) {
if (i == depth)
ofprofile << setw(10) << right << setprecision(2) << event->getPriority () << " ";
else
ofprofile << setw(10) << right << setprecision(2) << 0.0 << " ";
}
ofprofile << setw( 2) << right << event->getEventLevel() << endl;
}
}
Anabatic transient commit 18. Port of Kite (Katana), Yeah, Baby! Yeah! * Bug: In Hurricane, in StaticObservable::getObserver(), if the slot pointer is NULL, do not try to access the owner. Returns NULL, so the caller can be aware of the situation... * Change: In Hurricane, in BreakpointWidget & ExceptionWidget some cosmetic changes (fonts and window sizes). * Bug: In Anabatic, In AutoHorizontal::getConstraints(), take into account the constraints from the source AutoContact, as it holds the constraints transmitted by the RoutingPads and sets up by propageConstraintsFromRp(). It is likely to be a bug affecting the original Katabatic as well. * Change: In Anabatic, in RawGCellsUnder(), check that the segment is not completly oustside the cell abutment box and truncate the coordinates to the part that is inside. Use the "shrink" if we reach the east/north border. * Change: In Anabatic, in Configuration, no more decorator because we will use a true derived relationship. Katana *derives* from *Anabatic* and do not *decorate* it, so the Configuration can do the same. It also implies that we directly create a Katana engine, not an Anabatic one. * Change: In Anabatic, in Session, do not allow the opening of the Session in a standalone fashion (with a static method). Instead it must be opened using the relevant method of the Anabatic/Katana engine. This ensure we are opening the right Session type. * Change: In Anabatic, in AutoSegment_Aligneds() collection the seed segment is not part of the collection by default, but will be included if the Flags::WithSelf is set. * Change: In Configuration, all the flags value are now defined in two steps. Declared in the header and initialized in the module. This is to prevent the fact that on some cases, in relation with the Python "extern C" part modules, we need a true allocated variable. It was causing weird linking problems. A side effect is that they can no longer be used as entry is switches, have to replace them by if/else. * New: In Anabatic, new GCell::getNeighborAt() utility function. * Bug: In Anabatic, in GCell::doGrid(), tag all the GCells of the grid with the grid type... Back annote all the edges capacity (north & east) with the reserved local capacity. * New: Complete portage of Kite over Anabatic. The new engine is christened "Katana" for Kite-Analogic. When it's capabilities and performances will be on a part with Kite, it is to completly replace it (and take back the "Kite" name). Preliminary tests seems to show that, contrary to intuition (because built on a more complex/slower grid), it is even slightly faster than Kite 8-).
2016-08-15 09:30:13 -05:00
event->process( _eventQueue, _eventHistory, _eventLoop );
count++;
//if (count and not (count % 500)) {
// _pack( count, false );
//}
if (RoutingEvent::getProcesseds() >= limit) setInterrupt( true );
}
//_pack( count, true );
if (count and cmess2.enabled() and tty::enabled()) cmess1 << endl;
cdebug_log(9000,0) << "Deter| Repair Stage" << endl;
cmess1 << " o Repair Stage." << endl;
cdebug_log(159,0) << "Loadind Repair queue." << endl;
RoutingEvent::setStage( RoutingEvent::Repair );
for ( size_t i=0 ; (i<_eventHistory.size()) and not isInterrupted() ; i++ ) {
RoutingEvent* event = _eventHistory.getNth(i);
if (not event->isCloned() and event->isUnimplemented()) {
event->reschedule( _eventQueue, 0 );
}
}
_eventQueue.commit();
cmess2 << " <repair.queue:" << right << setw(8) << setfill('0')
<< _eventQueue.size() << ">" << endl;
count = 0;
//_eventQueue.prepareRepair();
while ( not _eventQueue.empty() and not isInterrupted() ) {
RoutingEvent* event = _eventQueue.pop();
if (tty::enabled()) {
cmess2 << " <repair.event:" << tty::bold << setw(8) << setfill('0')
<< RoutingEvent::getProcesseds() << tty::reset
<< " remains:" << right << setw(8) << setfill('0')
<< _eventQueue.size() << ">"
<< setfill(' ') << tty::reset << tty::cr;
cmess2.flush();
} else {
cmess2 << " <repair.event:" << setw(8) << setfill('0')
<< RoutingEvent::getProcesseds() << setfill(' ') << " "
<< event->getEventLevel() << ":" << event->getPriority() << "> "
<< event->getSegment()
<< endl;
cmess2.flush();
}
event->process( _eventQueue, _eventHistory, _eventLoop );
count++;
if (RoutingEvent::getProcesseds() >= limit ) setInterrupt( true );
}
if (count and cmess2.enabled() and tty::enabled()) cmess1 << endl;
size_t eventsCount = _eventHistory.size();
_eventHistory.clear();
_eventQueue.clear();
if (RoutingEvent::getAllocateds() > 0) {
cerr << Bug( "%d events remains after clear.", RoutingEvent::getAllocateds() ) << endl;
}
if (ofprofile.is_open()) ofprofile.close();
Anabatic transient commit 18. Port of Kite (Katana), Yeah, Baby! Yeah! * Bug: In Hurricane, in StaticObservable::getObserver(), if the slot pointer is NULL, do not try to access the owner. Returns NULL, so the caller can be aware of the situation... * Change: In Hurricane, in BreakpointWidget & ExceptionWidget some cosmetic changes (fonts and window sizes). * Bug: In Anabatic, In AutoHorizontal::getConstraints(), take into account the constraints from the source AutoContact, as it holds the constraints transmitted by the RoutingPads and sets up by propageConstraintsFromRp(). It is likely to be a bug affecting the original Katabatic as well. * Change: In Anabatic, in RawGCellsUnder(), check that the segment is not completly oustside the cell abutment box and truncate the coordinates to the part that is inside. Use the "shrink" if we reach the east/north border. * Change: In Anabatic, in Configuration, no more decorator because we will use a true derived relationship. Katana *derives* from *Anabatic* and do not *decorate* it, so the Configuration can do the same. It also implies that we directly create a Katana engine, not an Anabatic one. * Change: In Anabatic, in Session, do not allow the opening of the Session in a standalone fashion (with a static method). Instead it must be opened using the relevant method of the Anabatic/Katana engine. This ensure we are opening the right Session type. * Change: In Anabatic, in AutoSegment_Aligneds() collection the seed segment is not part of the collection by default, but will be included if the Flags::WithSelf is set. * Change: In Configuration, all the flags value are now defined in two steps. Declared in the header and initialized in the module. This is to prevent the fact that on some cases, in relation with the Python "extern C" part modules, we need a true allocated variable. It was causing weird linking problems. A side effect is that they can no longer be used as entry is switches, have to replace them by if/else. * New: In Anabatic, new GCell::getNeighborAt() utility function. * Bug: In Anabatic, in GCell::doGrid(), tag all the GCells of the grid with the grid type... Back annote all the edges capacity (north & east) with the reserved local capacity. * New: Complete portage of Kite over Anabatic. The new engine is christened "Katana" for Kite-Analogic. When it's capabilities and performances will be on a part with Kite, it is to completly replace it (and take back the "Kite" name). Preliminary tests seems to show that, contrary to intuition (because built on a more complex/slower grid), it is even slightly faster than Kite 8-).
2016-08-15 09:30:13 -05:00
_statistics.setEventsCount( eventsCount );
cdebug_tabw(159,-1);
return eventsCount;
}
void NegociateWindow::run ( Flags flags )
Anabatic transient commit 18. Port of Kite (Katana), Yeah, Baby! Yeah! * Bug: In Hurricane, in StaticObservable::getObserver(), if the slot pointer is NULL, do not try to access the owner. Returns NULL, so the caller can be aware of the situation... * Change: In Hurricane, in BreakpointWidget & ExceptionWidget some cosmetic changes (fonts and window sizes). * Bug: In Anabatic, In AutoHorizontal::getConstraints(), take into account the constraints from the source AutoContact, as it holds the constraints transmitted by the RoutingPads and sets up by propageConstraintsFromRp(). It is likely to be a bug affecting the original Katabatic as well. * Change: In Anabatic, in RawGCellsUnder(), check that the segment is not completly oustside the cell abutment box and truncate the coordinates to the part that is inside. Use the "shrink" if we reach the east/north border. * Change: In Anabatic, in Configuration, no more decorator because we will use a true derived relationship. Katana *derives* from *Anabatic* and do not *decorate* it, so the Configuration can do the same. It also implies that we directly create a Katana engine, not an Anabatic one. * Change: In Anabatic, in Session, do not allow the opening of the Session in a standalone fashion (with a static method). Instead it must be opened using the relevant method of the Anabatic/Katana engine. This ensure we are opening the right Session type. * Change: In Anabatic, in AutoSegment_Aligneds() collection the seed segment is not part of the collection by default, but will be included if the Flags::WithSelf is set. * Change: In Configuration, all the flags value are now defined in two steps. Declared in the header and initialized in the module. This is to prevent the fact that on some cases, in relation with the Python "extern C" part modules, we need a true allocated variable. It was causing weird linking problems. A side effect is that they can no longer be used as entry is switches, have to replace them by if/else. * New: In Anabatic, new GCell::getNeighborAt() utility function. * Bug: In Anabatic, in GCell::doGrid(), tag all the GCells of the grid with the grid type... Back annote all the edges capacity (north & east) with the reserved local capacity. * New: Complete portage of Kite over Anabatic. The new engine is christened "Katana" for Kite-Analogic. When it's capabilities and performances will be on a part with Kite, it is to completly replace it (and take back the "Kite" name). Preliminary tests seems to show that, contrary to intuition (because built on a more complex/slower grid), it is even slightly faster than Kite 8-).
2016-08-15 09:30:13 -05:00
{
cdebug_log(159,1) << "NegociateWindow::run()" << endl;
cmess1 << " o Running Negociate Algorithm" << endl;
TrackElement::setOverlapCostCB( NegociateOverlapCost );
RoutingEvent::resetProcesseds();
for ( size_t igcell=0 ; igcell<_gcells.size() ; ++igcell ) {
_createRouting( _gcells[igcell] );
}
Session::revalidate();
if (not (flags & Flags::PreRoutedStage)) {
_katana->preProcess();
Session::revalidate();
}
_katana->setMinimumWL( computeWirelength() );
#if defined(CHECK_DATABASE)
uint32_t overlaps = 0;
Anabatic transient commit 18. Port of Kite (Katana), Yeah, Baby! Yeah! * Bug: In Hurricane, in StaticObservable::getObserver(), if the slot pointer is NULL, do not try to access the owner. Returns NULL, so the caller can be aware of the situation... * Change: In Hurricane, in BreakpointWidget & ExceptionWidget some cosmetic changes (fonts and window sizes). * Bug: In Anabatic, In AutoHorizontal::getConstraints(), take into account the constraints from the source AutoContact, as it holds the constraints transmitted by the RoutingPads and sets up by propageConstraintsFromRp(). It is likely to be a bug affecting the original Katabatic as well. * Change: In Anabatic, in RawGCellsUnder(), check that the segment is not completly oustside the cell abutment box and truncate the coordinates to the part that is inside. Use the "shrink" if we reach the east/north border. * Change: In Anabatic, in Configuration, no more decorator because we will use a true derived relationship. Katana *derives* from *Anabatic* and do not *decorate* it, so the Configuration can do the same. It also implies that we directly create a Katana engine, not an Anabatic one. * Change: In Anabatic, in Session, do not allow the opening of the Session in a standalone fashion (with a static method). Instead it must be opened using the relevant method of the Anabatic/Katana engine. This ensure we are opening the right Session type. * Change: In Anabatic, in AutoSegment_Aligneds() collection the seed segment is not part of the collection by default, but will be included if the Flags::WithSelf is set. * Change: In Configuration, all the flags value are now defined in two steps. Declared in the header and initialized in the module. This is to prevent the fact that on some cases, in relation with the Python "extern C" part modules, we need a true allocated variable. It was causing weird linking problems. A side effect is that they can no longer be used as entry is switches, have to replace them by if/else. * New: In Anabatic, new GCell::getNeighborAt() utility function. * Bug: In Anabatic, in GCell::doGrid(), tag all the GCells of the grid with the grid type... Back annote all the edges capacity (north & east) with the reserved local capacity. * New: Complete portage of Kite over Anabatic. The new engine is christened "Katana" for Kite-Analogic. When it's capabilities and performances will be on a part with Kite, it is to completly replace it (and take back the "Kite" name). Preliminary tests seems to show that, contrary to intuition (because built on a more complex/slower grid), it is even slightly faster than Kite 8-).
2016-08-15 09:30:13 -05:00
Session::getKatanaEngine()->_check( overlaps, "after _createRouting(GCell*)" );
#endif
Apaired segments building for symmetric routing (step 1). * Change: In Hurricane::BaseFlags, store flags in uint64_t instead of unsigned int because we start to need more than 32 different flags in some tools. * New: In ::getString() & ::getRecord() templates, add support for std::array<>. * Change: In CRL::ToolEngine, add support for timer (time & memory measurements) displaced from Katabatic. This way all ToolEngine can use this feature. The _postCreate() method display the memory just after ToolEngine allocation. * Change: In Etesian::EtesianEngine, make use of the ToolEngine builtin timer (remove the local one). Forgot to call the base class _postCreate() and _preDestroy(). * Change: In Anabatic::AnabaticEngine, make use of the ToolEngine builtin timer (remove the local one). * New: In Anabatic, new AutoSegments_Connecteds() collection. This Collection allows a deterministic walkthough *all* the AutoSegments connected either to source or target of one AutoSegment. * New: In Anabatic::AutoContactTerminal::isEndPoint() to check if an AutoContactTerminal is the *only one* anchored on a RoutingPad, thus being a true "end point" and not a kind of feed-through. * New: In Katana::KatanaEngine, added support for symmetric nets. Created new class DataSymmetric to store symmetric information of a net (mainly the paired AutoSegments). Added KatanaEngine::runSymmetricRouter(), for now only build the DataSymmetric informations. More to come... * Change: In Katana::GraphicKatanaEngine::_runTest(), now perform symmetric information building the non-symmetric routing.
2017-03-12 13:34:12 -05:00
if (flags & Flags::SymmetricStage) {
_katana->runSymmetricRouter();
Session::revalidate();
}
Anabatic transient commit 18. Port of Kite (Katana), Yeah, Baby! Yeah! * Bug: In Hurricane, in StaticObservable::getObserver(), if the slot pointer is NULL, do not try to access the owner. Returns NULL, so the caller can be aware of the situation... * Change: In Hurricane, in BreakpointWidget & ExceptionWidget some cosmetic changes (fonts and window sizes). * Bug: In Anabatic, In AutoHorizontal::getConstraints(), take into account the constraints from the source AutoContact, as it holds the constraints transmitted by the RoutingPads and sets up by propageConstraintsFromRp(). It is likely to be a bug affecting the original Katabatic as well. * Change: In Anabatic, in RawGCellsUnder(), check that the segment is not completly oustside the cell abutment box and truncate the coordinates to the part that is inside. Use the "shrink" if we reach the east/north border. * Change: In Anabatic, in Configuration, no more decorator because we will use a true derived relationship. Katana *derives* from *Anabatic* and do not *decorate* it, so the Configuration can do the same. It also implies that we directly create a Katana engine, not an Anabatic one. * Change: In Anabatic, in Session, do not allow the opening of the Session in a standalone fashion (with a static method). Instead it must be opened using the relevant method of the Anabatic/Katana engine. This ensure we are opening the right Session type. * Change: In Anabatic, in AutoSegment_Aligneds() collection the seed segment is not part of the collection by default, but will be included if the Flags::WithSelf is set. * Change: In Configuration, all the flags value are now defined in two steps. Declared in the header and initialized in the module. This is to prevent the fact that on some cases, in relation with the Python "extern C" part modules, we need a true allocated variable. It was causing weird linking problems. A side effect is that they can no longer be used as entry is switches, have to replace them by if/else. * New: In Anabatic, new GCell::getNeighborAt() utility function. * Bug: In Anabatic, in GCell::doGrid(), tag all the GCells of the grid with the grid type... Back annote all the edges capacity (north & east) with the reserved local capacity. * New: Complete portage of Kite over Anabatic. The new engine is christened "Katana" for Kite-Analogic. When it's capabilities and performances will be on a part with Kite, it is to completly replace it (and take back the "Kite" name). Preliminary tests seems to show that, contrary to intuition (because built on a more complex/slower grid), it is even slightly faster than Kite 8-).
2016-08-15 09:30:13 -05:00
_flags |= flags;
_negociate();
printStatistics();
if (flags & Flags::PreRoutedStage) {
_katana->setFixedPreRouted();
}
Session::revalidate();
Session::get()->isEmpty();
# if defined(CHECK_DATABASE)
_katana->_check( overlaps, "after negociation" );
# endif
cdebug_tabw(159,-1);
}
void NegociateWindow::printStatistics () const
{
cmess1 << " o Computing statistics." << endl;
cmess1 << Dots::asSizet(" - Processeds Events Total",RoutingEvent::getProcesseds()) << endl;
cmess1 << Dots::asSizet(" - Unique Events Total"
,(RoutingEvent::getProcesseds() - RoutingEvent::getCloneds())) << endl;
cmess1 << Dots::asSizet(" - # of GCells",_statistics.getGCellsCount()) << endl;
_katana->printCompletion();
addMeasure<size_t>( getCell(), "Events" , RoutingEvent::getProcesseds(), 12 );
addMeasure<size_t>( getCell(), "UEvents", RoutingEvent::getProcesseds()-RoutingEvent::getCloneds(), 12 );
Histogram* densityHistogram = new Histogram ( 1.0, 0.1, 2 );
addMeasure<Histogram>( getCell(), "GCells Density Histogram", densityHistogram );
densityHistogram->setFileExtension( ".density.histogram" );
densityHistogram->setMainTitle ( "GCell Densities" );
densityHistogram->setTitle ( "Avg. Density", 0 );
densityHistogram->setTitle ( "Peak Density", 1 );
densityHistogram->setColor ( "green", 0 );
densityHistogram->setColor ( "red" , 1 );
const vector<GCell*>& gcells = getKatanaEngine()->getGCells();
getKatanaEngine()->setDensityMode( Anabatic::AnabaticEngine::MaxHVDensity );
for ( size_t igcell=0 ; igcell<gcells.size() ; ++igcell ) {
densityHistogram->addSample( gcells[igcell]->getDensity(), 0 );
}
getKatanaEngine()->setDensityMode( Anabatic::AnabaticEngine::MaxDensity );
for ( size_t igcell=0 ; igcell<gcells.size() ; ++igcell ) {
densityHistogram->addSample( gcells[igcell]->getDensity(), 1 );
}
densityHistogram->normalize( 0 );
densityHistogram->normalize( 1 );
}
string NegociateWindow::_getString () const
{
ostringstream os;
os << "<" << _getTypeName() << ">";
return os.str();
}
Record* NegociateWindow::_getRecord () const
{
Record* record = new Record ( _getString() );
record->add( getSlot( "_gcells", _gcells ) );
return record;
}
} // Katana namespace.