coriolis/kite/src/KiteEngine.cpp

845 lines
29 KiB
C++
Raw Normal View History

// -*- C++ -*-
//
// This file is part of the Coriolis Software.
Improved UpdateSession & exception catching. Start of RoutingGauge implem. 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.
2014-04-20 12:25:08 -05:00
// Copyright (c) UPMC 2008-2014, 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 : "./KiteEngine.cpp" |
// +-----------------------------------------------------------------+
#include <sstream>
#include <fstream>
#include <iomanip>
#include "hurricane/DebugSession.h"
#include "hurricane/Bug.h"
#include "hurricane/Error.h"
#include "hurricane/Warning.h"
#include "hurricane/Breakpoint.h"
#include "hurricane/Layer.h"
#include "hurricane/Net.h"
#include "hurricane/Pad.h"
#include "hurricane/Plug.h"
#include "hurricane/Cell.h"
#include "hurricane/Instance.h"
#include "hurricane/Vertical.h"
#include "hurricane/Horizontal.h"
#include "crlcore/Measures.h"
#include "knik/Vertex.h"
#include "knik/Edge.h"
#include "knik/Graph.h"
#include "knik/KnikEngine.h"
#include "katabatic/AutoContact.h"
#include "katabatic/GCellGrid.h"
#include "kite/DataNegociate.h"
#include "kite/RoutingPlane.h"
#include "kite/Session.h"
#include "kite/TrackSegment.h"
#include "kite/NegociateWindow.h"
#include "kite/KiteEngine.h"
namespace Kite {
using std::cout;
using std::cerr;
using std::endl;
using std::setw;
using std::left;
using std::ostream;
using std::ofstream;
using std::ostringstream;
using std::setprecision;
using std::vector;
using Hurricane::DebugSession;
using Hurricane::tab;
using Hurricane::inltrace;
using Hurricane::ltracein;
using Hurricane::ltraceout;
using Hurricane::ForEachIterator;
using Hurricane::Bug;
using Hurricane::Error;
using Hurricane::Warning;
using Hurricane::Breakpoint;
using Hurricane::Box;
* ./kite: - 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.
2010-12-04 09:25:48 -06:00
using Hurricane::Torus;
using Hurricane::Layer;
using Hurricane::Cell;
using CRL::addMeasure;
using CRL::Measures;
using CRL::MeasuresSet;
using Knik::KnikEngine;
using Katabatic::AutoContact;
using Katabatic::AutoSegmentLut;
* ./kite: - 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.
2010-12-04 09:25:48 -06:00
using Katabatic::ChipTools;
const char* missingRW =
"%s :\n\n"
" Cell %s do not have any KiteEngine (or not yet created).\n";
// -------------------------------------------------------------------
// Class : "Kite::KiteEngine".
Name KiteEngine::_toolName = "Kite";
const Name& KiteEngine::staticGetName ()
{ return _toolName; }
KiteEngine* KiteEngine::get ( const Cell* cell )
{ return static_cast<KiteEngine*>(ToolEngine::get(cell,staticGetName())); }
KiteEngine::KiteEngine ( Cell* cell )
: KatabaticEngine (cell)
Improved UpdateSession & exception catching. Start of RoutingGauge implem. 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.
2014-04-20 12:25:08 -05:00
, _viewer (NULL)
, _knik (NULL)
, _blockageNet (NULL)
, _configuration (new Configuration(getKatabaticConfiguration()))
, _routingPlanes ()
, _negociateWindow(NULL)
, _minimumWL (0.0)
, _toolSuccess (false)
{ }
void KiteEngine::_postCreate ()
{
KatabaticEngine::_postCreate ();
#ifdef KNIK_NOT_EMBEDDED
size_t maxDepth = getRoutingGauge()->getDepth();
_routingPlanes.reserve( maxDepth );
for ( size_t depth=0 ; depth < maxDepth ; depth++ ) {
_routingPlanes.push_back( RoutingPlane::create( this, depth ) );
}
#endif
}
KiteEngine* KiteEngine::create ( Cell* cell )
{
KiteEngine* kite = new KiteEngine ( cell );
kite->_postCreate();
return kite;
}
void KiteEngine::_preDestroy ()
{
ltrace(90) << "KiteEngine::_preDestroy()" << endl;
ltracein(90);
cmess1 << " o Deleting ToolEngine<" << getName() << "> from Cell <"
<< _cell->getName() << ">" << endl;
if (getState() < Katabatic::EngineGutted)
setState( Katabatic::EnginePreDestroying );
_gutKite();
KatabaticEngine::_preDestroy();
cmess2 << " - RoutingEvents := " << RoutingEvent::getAllocateds() << endl;
if (not ToolEngine::inDestroyAll()) {
KnikEngine* attachedKnik = KnikEngine::get( getCell() );
if (_knik != attachedKnik) {
cerr << Error("Knik attribute differs from the Cell attached one (must be the same)\n"
" On: <%s>."
,getString(getCell()->getName()).c_str()) << endl;
_knik = attachedKnik;
}
_knik->destroy();
}
ltraceout(90);
}
KiteEngine::~KiteEngine ()
{ delete _configuration; }
const Name& KiteEngine::getName () const
{ return _toolName; }
Configuration* KiteEngine::getConfiguration ()
{ return _configuration; }
unsigned int KiteEngine::getRipupLimit ( const TrackElement* segment ) const
{
if (segment->isBlockage()) return 0;
if (segment->isStrap ()) return _configuration->getRipupLimit( Configuration::StrapRipupLimit );
if (segment->isGlobal()) {
Katabatic::GCellVector gcells;
segment->getGCells( gcells );
if (gcells.size() > 2)
return _configuration->getRipupLimit( Configuration::LongGlobalRipupLimit );
return _configuration->getRipupLimit( Configuration::GlobalRipupLimit );
}
return _configuration->getRipupLimit( Configuration::LocalRipupLimit );
}
RoutingPlane* KiteEngine::getRoutingPlaneByIndex ( size_t index ) const
{
if (index >= getRoutingPlanesSize() ) return NULL;
return _routingPlanes[index];
}
RoutingPlane* KiteEngine::getRoutingPlaneByLayer ( const Layer* layer ) const
{
for ( size_t index=0 ; index < getRoutingPlanesSize() ; index++ ) {
if (_routingPlanes[index]->getLayer() == layer)
return _routingPlanes[index];
}
return NULL;
}
Track* KiteEngine::getTrackByPosition ( const Layer* layer, DbU::Unit axis, unsigned int mode ) const
{
RoutingPlane* plane = getRoutingPlaneByLayer( layer );
if (not plane) return NULL;
return plane->getTrackByPosition( axis, mode );
}
void KiteEngine::setInterrupt ( bool state )
{
if (_negociateWindow) {
_negociateWindow->setInterrupt( state );
cerr << "Interrupt [CRTL+C] of " << this << endl;
}
}
void KiteEngine::createGlobalGraph ( unsigned int mode )
{
* ./kite: - 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.
2010-12-04 09:25:48 -06:00
Cell* cell = getCell();
//Box cellBb = cell->getBoundingBox();
if (not _knik) {
unsigned int flags = Cell::WarnOnUnplacedInstances;
flags |= (mode & KtBuildGlobalRouting) ? Cell::BuildRings : 0;
cell->flattenNets( flags );
Improved UpdateSession & exception catching. Start of RoutingGauge implem. 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.
2014-04-20 12:25:08 -05:00
// Test signals from <snx2013>.
//DebugSession::addToTrace( getCell(), "core.snx_inst.a2_x2_8_sig" );
//DebugSession::addToTrace( getCell(), "m_clock" );
//DebugSession::addToTrace( getCell(), "a2_x2_8_sig" );
KatabaticEngine::chipPrep();
KnikEngine::setHEdgeReservedLocal( 0 );
KnikEngine::setVEdgeReservedLocal( 0 );
_knik = KnikEngine::create( cell
, 1 // _congestion
, 2 // _preCongestion
, false // _benchMode
, true // _useSegments
, 2.5 // _edgeCost
);
Improved UpdateSession & exception catching. Start of RoutingGauge implem. 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.
2014-04-20 12:25:08 -05:00
_knik->setRoutingGauge( getConfiguration()->getRoutingGauge() );
_knik->setAllowedDepth( getConfiguration()->getAllowedDepth() );
_knik->createRoutingGraph();
KnikEngine::setHEdgeReservedLocal( getHTracksReservedLocal() );
KnikEngine::setVEdgeReservedLocal( getVTracksReservedLocal() );
* ./kite: - 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.
2010-12-04 09:25:48 -06:00
// Decrease the edge's capacity only under the core area.
const ChipTools& chipTools = getChipTools();
size_t coreReserved = 0;
size_t coronaReserved = 4;
* ./kite: - 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.
2010-12-04 09:25:48 -06:00
forEach ( Knik::Vertex*, ivertex, _knik->getRoutingGraph()->getVertexes() ) {
for ( int i=0 ; i<2 ; ++i ) {
* ./kite: - 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.
2011-01-25 11:16:50 -06:00
Knik::Edge* edge = NULL;
if (i==0) {
* ./kite: - 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.
2010-12-04 09:25:48 -06:00
edge = ivertex->getHEdgeOut();
if (not edge) continue;
Improved UpdateSession & exception catching. Start of RoutingGauge implem. 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.
2014-04-20 12:25:08 -05:00
if (chipTools.hPadsEnclosed(edge->getBoundingBox())) {
edge->setCapacity( 0 );
* ./kite: - 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.
2010-12-04 09:25:48 -06:00
continue;
}
coreReserved = getHTracksReservedLocal();
* ./kite: - 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.
2010-12-04 09:25:48 -06:00
} else {
edge = ivertex->getVEdgeOut();
if (not edge) continue;
Improved UpdateSession & exception catching. Start of RoutingGauge implem. 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.
2014-04-20 12:25:08 -05:00
if (chipTools.vPadsEnclosed(edge->getBoundingBox())) {
edge->setCapacity( 0 );
* ./kite: - 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.
2010-12-04 09:25:48 -06:00
continue;
}
coreReserved = getVTracksReservedLocal();
* ./kite: - 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.
2010-12-04 09:25:48 -06:00
}
size_t edgeReserved = 0;
if (chipTools.getCorona().getInnerBox().contains(edge->getBoundingBox())) {
edgeReserved = coreReserved;
} else if (chipTools.getCorona().getOuterBox().contains(edge->getBoundingBox())) {
edgeReserved = coronaReserved;
* ./kite: - 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.
2011-01-25 11:16:50 -06:00
}
size_t capacity = (edge->getCapacity()>edgeReserved)
? (edge->getCapacity()-edgeReserved) : 0;
Improved UpdateSession & exception catching. Start of RoutingGauge implem. 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.
2014-04-20 12:25:08 -05:00
//cerr << "Appling capacity percentage " << (edgePercent*100.0) << "% ("
// << capacity << ") on: " << edge << endl;
edge->setCapacity( capacity );
* ./kite: - 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.
2010-12-04 09:25:48 -06:00
}
}
}
}
void KiteEngine::createDetailedGrid ()
{
KatabaticEngine::createDetailedGrid();
size_t maxDepth = getRoutingGauge()->getDepth();
_routingPlanes.reserve( maxDepth );
for ( size_t depth=0 ; depth < maxDepth ; depth++ ) {
_routingPlanes.push_back( RoutingPlane::create ( this, depth ) );
}
}
void KiteEngine::saveGlobalSolution ()
{
if (getState() < Katabatic::EngineGlobalLoaded)
throw Error ("KiteEngine::saveGlobalSolution(): Global routing not present yet.");
if (getState() > Katabatic::EngineGlobalLoaded)
throw Error ("KiteEngine::saveGlobalSolution(): Cannot save after detailed routing.");
_knik->saveSolution();
}
void KiteEngine::annotateGlobalGraph ()
{
cmess1 << " o Back annotate global routing graph." << endl;
* ./kite: - 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.
2010-12-04 09:25:48 -06:00
const Torus& chipCorona = getChipTools().getCorona();
int hEdgeCapacity = 0;
int vEdgeCapacity = 0;
for ( size_t depth=0 ; depth<_routingPlanes.size() ; ++depth ) {
RoutingPlane* rp = _routingPlanes[depth];
Improved UpdateSession & exception catching. Start of RoutingGauge implem. 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.
2014-04-20 12:25:08 -05:00
if (rp->getLayerGauge()->getType() == Constant::PinOnly ) continue;
if (rp->getLayerGauge()->getDepth() > getConfiguration()->getAllowedDepth() ) continue;
* ./kite: - 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.
2010-12-04 09:25:48 -06:00
if (rp->getDirection() == KbHorizontal) ++hEdgeCapacity;
* ./kite: - 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.
2010-12-04 09:25:48 -06:00
else ++vEdgeCapacity;
}
for ( size_t depth=0 ; depth<_routingPlanes.size() ; ++depth ) {
RoutingPlane* rp = _routingPlanes[depth];
if (rp->getLayerGauge()->getType() == Constant::PinOnly ) continue;
Improved UpdateSession & exception catching. Start of RoutingGauge implem. 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.
2014-04-20 12:25:08 -05:00
if (rp->getLayerGauge()->getDepth() > getConfiguration()->getAllowedDepth() ) continue;
size_t tracksSize = rp->getTracksSize();
for ( size_t itrack=0 ; itrack<tracksSize ; ++itrack ) {
Track* track = rp->getTrackByIndex ( itrack );
* ./kite: - 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.
2010-12-04 09:25:48 -06:00
ltrace(300) << "Capacity from: " << track << endl;
if (track->getDirection() == KbHorizontal) {
for ( size_t ielement=0 ; ielement<track->getSize() ; ++ielement ) {
TrackElement* element = track->getSegment( ielement );
if (element->getNet() == NULL) {
ltrace(300) << "Reject capacity from (not Net): " << element << endl;
continue;
}
* ./kite: - 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.
2010-12-04 09:25:48 -06:00
if ( (not element->isFixed()) and (not element->isBlockage()) ) {
ltrace(300) << "Reject capacity from (neither fixed nor blockage): " << element << endl;
continue;
}
* ./kite: - 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.
2010-12-04 09:25:48 -06:00
Box elementBb = element->getBoundingBox();
int elementCapacity = (chipCorona.contains(elementBb)) ? -hEdgeCapacity : -1;
ltrace(300) << "Capacity from: " << element << ":" << elementCapacity << endl;
Katabatic::GCell* gcell = getGCellGrid()->getGCell( Point(element->getSourceU(),track->getAxis()) );
Katabatic::GCell* end = getGCellGrid()->getGCell( Point(element->getTargetU(),track->getAxis()) );
Katabatic::GCell* right = NULL;
if (not gcell) {
cerr << Warning("annotageGlobalGraph(): TrackElement outside GCell grid.") << endl;
continue;
}
* ./kite: - 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.
2010-12-04 09:25:48 -06:00
while ( gcell and (gcell != end) ) {
right = gcell->getRight();
if (right == NULL) break;
_knik->increaseEdgeCapacity( gcell->getColumn()
, gcell->getRow()
, right->getColumn()
, right->getRow()
, elementCapacity );
gcell = right;
}
}
} else {
for ( size_t ielement=0 ; ielement<track->getSize() ; ++ielement ) {
TrackElement* element = track->getSegment( ielement );
if (element->getNet() == NULL) {
ltrace(300) << "Reject capacity from (not Net): " << element << endl;
continue;
}
* ./kite: - 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.
2010-12-04 09:25:48 -06:00
if ( (not element->isFixed()) and not (element->isBlockage()) ) {
ltrace(300) << "Reject capacity from (neither fixed nor blockage): " << element << endl;
continue;
}
* ./kite: - 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.
2010-12-04 09:25:48 -06:00
Box elementBb = element->getBoundingBox();
int elementCapacity = (chipCorona.contains(elementBb)) ? -vEdgeCapacity : -1;
ltrace(300) << "Capacity from: " << element << ":" << elementCapacity << endl;
Katabatic::GCell* gcell = getGCellGrid()->getGCell( Point(track->getAxis(),element->getSourceU()) );
Katabatic::GCell* end = getGCellGrid()->getGCell( Point(track->getAxis(),element->getTargetU()) );
Katabatic::GCell* up = NULL;
if (not gcell) {
cerr << Warning("annotageGlobalGraph(): TrackElement outside GCell grid.") << endl;
continue;
}
while ( gcell and (gcell != end) ) {
up = gcell->getUp();
if (up == NULL) break;
_knik->increaseEdgeCapacity( gcell->getColumn()
, gcell->getRow()
, up->getColumn()
, up->getRow()
, elementCapacity );
gcell = up;
}
}
}
}
}
}
void KiteEngine::runGlobalRouter ( unsigned int mode )
{
if (getState() >= Katabatic::EngineGlobalLoaded)
throw Error ("KiteEngine::runGlobalRouter(): Global routing already done or loaded.");
Session::open( this );
createGlobalGraph( mode );
ExtensionCap support and source/target terminal flags in Katabatic & Kite. 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.
2014-05-25 08:00:35 -05:00
// Test signals from <multi4_a>.
//DebugSession::addToTrace( getCell(), "aux34" );
// Test signals from <addaccu>.
//DebugSession::addToTrace( getCell(), "auxsc37" );
// Test signals from <amd2901_core_flat>.
//DebugSession::addToTrace( getCell(), "cout" );
//DebugSession::addToTrace( getCell(), "acc_reg_ckx" );
//DebugSession::addToTrace( getCell(), "acc_reg_nckx" );
//DebugSession::addToTrace( getCell(), "i(0)" );
//DebugSession::addToTrace( getCell(), "ram_nmux_0_sel0" );
//DebugSession::addToTrace( getCell(), "ram_adrb_14" );
//DebugSession::addToTrace( getCell(), "ram_adrb_9" );
//DebugSession::addToTrace( getCell(), "ram_adra(11)" );
//DebugSession::addToTrace( getCell(), "ram_adra(7)" );
//DebugSession::addToTrace( getCell(), "ram_adrb(8)" );
//DebugSession::addToTrace( getCell(), "alu_carry(1)" );
//DebugSession::addToTrace( getCell(), "alu_np(0)" );
//DebugSession::addToTrace( getCell(), "ram_q2(0)" );
//DebugSession::addToTrace( getCell(), "ram_q1(0)" );
//DebugSession::addToTrace( getCell(), "ram_i_up" );
// Test signals from <amd2901> (M1-VLSI).
//DebugSession::addToTrace( getCell(), "zero_to_pads" );
//DebugSession::addToTrace( getCell(), "shift_r" );
//DebugSession::addToTrace( getCell(), "cin_from_pads" );
// Test signals from <MIPS> (R3000,micro-programmed).
//DebugSession::addToTrace( getCell(), "scout" );
//DebugSession::addToTrace( getCell(), "adr_1_n" );
//DebugSession::addToTrace( getCell(), "codop_18" );
//DebugSession::addToTrace( getCell(), "frz_ctl(10)" );
//DebugSession::addToTrace( getCell(), "ctl_seq_mbk_not_ep_80" );
//DebugSession::addToTrace( getCell(), "ctl_sts_mbk_not_ctlrw_in_2" );
//DebugSession::addToTrace( getCell(), "dpt_wm_rf_adr4x" );
//DebugSession::addToTrace( getCell(), "crsrin_1" );
//DebugSession::addToTrace( getCell(), "ctl_seq_oa2ao222_x2_2" );
//DebugSession::addToTrace( getCell(), "dpt_ishifter_muxoutput_81" );
//DebugSession::addToTrace( getCell(), "ctl_seq_mbk_not_ep_7" );
//DebugSession::addToTrace( getCell(), "ctl_sts_mbk_not_adel_r" );
//DebugSession::addToTrace( getCell(), "ctl_seq_no2_x1_88" );
//DebugSession::addToTrace( getCell(), "ctl_seq_ep_31" );
//DebugSession::addToTrace( getCell(), "dpt_opyir16ins_mxn1" );
//DebugSession::addToTrace( getCell(), "dpt_ishifter_muxoutput_132" );
// Test signals from <MIPS> (R3000,pipeline).
//DebugSession::addToTrace( getCell(), "nb(0)" );
//DebugSession::addToTrace( getCell(), "mips_r3000_1m_dp_mux32_data_e_sm_sel0" );
//DebugSession::addToTrace( getCell(), "mips_r3000_1m_dp_lo_rw(16)" );
//DebugSession::addToTrace( getCell(), "mips_r3000_1m_dp_hi_rw(27)" );
//DebugSession::addToTrace( getCell(), "mips_r3000_1m_dp_data_e_sm(25)" );
//DebugSession::addToTrace( getCell(), "mips_r3000_1m_dp_nul_s_eq_z_sd_nul_3" );
//DebugSession::addToTrace( getCell(), "mips_r3000_1m_dp_shift32_rshift_se_muxoutput(143)" );
//DebugSession::addToTrace( getCell(), "rsdnbr_sd(19)" );
//DebugSession::addToTrace( getCell(), "mips_r3000_1m_dp_res_se(14)" );
//DebugSession::addToTrace( getCell(), "mips_r3000_1m_dp_res_re(0)" );
//DebugSession::addToTrace( getCell(), "wreg_sw(1)" );
//DebugSession::addToTrace( getCell(), "mips_r3000_1m_dp_shift32_rshift_se_msb" );
//DebugSession::addToTrace( getCell(), "mips_r3000_1m_ct_mx2_x2_2_sig" );
//DebugSession::addToTrace( getCell(), "mips_r3000_1m_dp_mux32_s_mw_se_sel0" );
//DebugSession::addToTrace( getCell(), "mips_r3000_1m_dp_mux32_badr_sd_sel1" );
//DebugSession::addToTrace( getCell(), "mips_r3000_1m_dp_addsub32_carith_se_pi_3_21" );
//DebugSession::addToTrace( getCell(), "mips_r3000_1m_dp_addsub32_carith_se_pi_4_26" );
//DebugSession::addToTrace( getCell(), "mips_r3000_1m_ct_cause_rx(1)" );
Improved UpdateSession & exception catching. Start of RoutingGauge implem. 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.
2014-04-20 12:25:08 -05:00
// Test signals from <MIPS> (R3000,pipeline+chip).
//DebugSession::addToTrace( getCell(), "mips_r3000_core.mips_r3000_1m_dp.banc.reada0" );
//DebugSession::addToTrace( getCell(), "mips_r3000_core.mips_r3000_1m_ct.i_ri(29)" );
//DebugSession::addToTrace( getCell(), "mips_r3000_core.mips_r3000_1m_ct.not_opcod_re(4)" );
//DebugSession::addToTrace( getCell(), "d_out_i(10)" );
//DebugSession::addToTrace( getCell(), "dout_e_i(0)" );
//DebugSession::addToTrace( getCell(), "dout_e_i(1)" );
//DebugSession::addToTrace( getCell(), "dout_e_i(2)" );
//DebugSession::addToTrace( getCell(), "i_ack_i" );
//DebugSession::addToTrace( getCell(), "mips_r3000_core.mips_r3000_1m_dp.data_rm(7)" );
Improved UpdateSession & exception catching. Start of RoutingGauge implem. 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.
2014-04-20 12:25:08 -05:00
// Test signals from <snx2013>.
//DebugSession::addToTrace( getCell(), "core.snx_inst.not_v_inc_out(9)" );
createDetailedGrid();
buildPowerRails();
protectRoutingPads();
Session::revalidate();
if (mode & KtLoadGlobalRouting) {
_knik->loadSolution();
} else {
annotateGlobalGraph();
_knik->run();
}
setState( Katabatic::EngineGlobalLoaded );
Session::close();
}
void KiteEngine::loadGlobalRouting ( unsigned int method, KatabaticEngine::NetSet& nets )
{
KatabaticEngine::loadGlobalRouting( method, nets );
Session::open( this );
getGCellGrid()->checkEdgeOverflow( getHTracksReservedLocal(), getVTracksReservedLocal() );
Session::close();
}
void KiteEngine::runNegociate ( unsigned int slowMotion )
{
if (_negociateWindow) return;
startMeasures();
Session::open( this );
_negociateWindow = NegociateWindow::create( this );
_negociateWindow->setGCells( *(getGCellGrid()->getGCellVector()) );
_computeCagedConstraints();
_negociateWindow->run( slowMotion );
_negociateWindow->printStatistics();
_negociateWindow->destroy();
_negociateWindow = NULL;
Session::close();
stopMeasures();
//if ( _editor ) _editor->refresh ();
printMeasures( "algo" );
Session::open( this );
unsigned int overlaps = 0;
size_t hTracksReservedLocal = getHTracksReservedLocal();
size_t vTracksReservedLocal = getVTracksReservedLocal();
KnikEngine* knik = KnikEngine::get( getCell() );
if (knik) {
hTracksReservedLocal = knik->getHEdgeReservedLocal();
vTracksReservedLocal = knik->getVEdgeReservedLocal();
}
if (cparanoid.enabled()) {
cparanoid << " o Post-checking Knik capacity overload h:" << hTracksReservedLocal
<< " v:." << vTracksReservedLocal << endl;
getGCellGrid()->checkEdgeOverflow( hTracksReservedLocal, vTracksReservedLocal );
}
_check( overlaps );
Session::close();
_toolSuccess = _toolSuccess and (overlaps == 0);
}
void KiteEngine::printCompletion () const
{
size_t routeds = 0;
unsigned long long totalWireLength = 0;
unsigned long long routedWireLength = 0;
vector<TrackElement*> unrouteds;
ostringstream result;
AutoSegmentLut::const_iterator ilut = _getAutoSegmentLut().begin();
for ( ; ilut != _getAutoSegmentLut().end() ; ilut++ ) {
TrackElement* segment = _lookup( ilut->second );
if (segment == NULL) continue;
unsigned long long wl = (unsigned long long)DbU::toLambda( segment->getLength() );
if (wl > 100000) {
cerr << Error("KiteEngine::printCompletion(): Suspiciously long wire: %llu for %p:%s"
,wl,ilut->first,getString(segment).c_str()) << endl;
continue;
}
if (segment->isFixed() or segment->isBlockage()) continue;
// if (segment->isSameLayerDogleg()) {
// cerr << " Same layer:" << segment << endl;
// cerr << " S: " << segment->base()->getAutoSource() << endl;
// cerr << " T: " << segment->base()->getAutoTarget() << endl;
// }
totalWireLength += wl;
if (segment->getTrack() != NULL) {
routeds++;
routedWireLength += wl;
} else {
unrouteds.push_back( segment );
}
}
float segmentRatio = (float)(routeds) / (float)(routeds+unrouteds.size()) * 100.0;
float wireLengthRatio = (float)(routedWireLength) / (float)(totalWireLength) * 100.0;
result << setprecision(4) << segmentRatio
<< "% [" << routeds << "+" << unrouteds.size() << "]";
cmess1 << Dots::asString( " - Track Segment Completion Ratio", result.str() ) << endl;
result.str("");
result << setprecision(4) << wireLengthRatio
<< "% [" << totalWireLength << "+"
<< (totalWireLength - routedWireLength) << "]";
cmess1 << Dots::asString( " - Wire Length Completion Ratio", result.str() ) << endl;
float expandRatio = 1.0;
if (_minimumWL != 0.0) {
* ./kite: - 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.
2011-01-25 11:16:50 -06:00
expandRatio = ((totalWireLength-_minimumWL) / _minimumWL) * 100.0;
result.str("");
result << setprecision(3) << expandRatio << "% [min:" << setprecision(9) << _minimumWL << "]";
cmess1 << Dots::asString( " - Wire Length Expand Ratio", result.str() ) << endl;
}
_toolSuccess = (unrouteds.empty());
if (not unrouteds.empty()) {
cerr << " o Routing did not complete, unrouted segments:" << endl;
for ( size_t i=0; i<unrouteds.size() ; ++i ) {
cerr << " " << setw(4) << (i+1) << "| " << unrouteds[i] << endl;
}
}
addMeasure<size_t> ( getCell(), "Segs" , routeds+unrouteds.size() );
addMeasure<unsigned long long>( getCell(), "DWL(l)" , totalWireLength , 12 );
addMeasure<unsigned long long>( getCell(), "fWL(l)" , totalWireLength-routedWireLength , 12 );
addMeasure<double> ( getCell(), "WLER(%)", (expandRatio-1.0)*100.0 );
}
void KiteEngine::dumpMeasures ( ostream& out ) const
{
vector<Name> measuresLabels;
measuresLabels.push_back( "Gates" );
measuresLabels.push_back( "GCells" );
measuresLabels.push_back( "knikT" );
measuresLabels.push_back( "knikS" );
measuresLabels.push_back( "GWL(l)" );
measuresLabels.push_back( "Area(l2)");
measuresLabels.push_back( "Sat." );
measuresLabels.push_back( "loadT" );
measuresLabels.push_back( "loadS" );
measuresLabels.push_back( "Globals" );
measuresLabels.push_back( "Edges" );
measuresLabels.push_back( "assignT" );
measuresLabels.push_back( "algoT" );
measuresLabels.push_back( "algoS" );
measuresLabels.push_back( "finT" );
measuresLabels.push_back( "Segs" );
measuresLabels.push_back( "DWL(l)" );
measuresLabels.push_back( "fWL(l)" );
measuresLabels.push_back( "WLER(%)" );
measuresLabels.push_back( "Events" );
measuresLabels.push_back( "UEvents" );
const MeasuresSet* measures = Measures::get( getCell() );
out << "#" << endl;
out << "# " << getCell()->getName() << endl;
out << measures->toStringHeaders(measuresLabels) << endl;
out << measures->toStringDatas (measuresLabels) << endl;
measures->toGnuplot( "GCells Density Histogram", getString(getCell()->getName()) );
}
void KiteEngine::dumpMeasures () const
{
ostringstream path;
path << getCell()->getName() << ".knik-kite.dat";
ofstream sfile ( path.str().c_str() );
dumpMeasures( sfile );
sfile.close();
}
bool KiteEngine::_check ( unsigned int& overlap, const char* message ) const
{
cmess1 << " o Checking Kite Database coherency." << endl;
bool coherency = true;
coherency = coherency and KatabaticEngine::_check( message );
for ( size_t i=0 ; i<_routingPlanes.size() ; i++ )
* ./Kite: - 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.
2010-12-30 12:42:17 -06:00
coherency = _routingPlanes[i]->_check(overlap) and coherency;
Katabatic::Session* ktbtSession = Session::base ();
forEach ( Net*, inet, getCell()->getNets() ) {
forEach ( Segment*, isegment, inet->getComponents().getSubSet<Segment*>() ) {
AutoSegment* autoSegment = ktbtSession->lookup( *isegment );
if (not autoSegment) continue;
if (not autoSegment->isCanonical()) continue;
TrackElement* trackSegment = Session::lookup( *isegment );
if (not trackSegment) {
coherency = false;
cerr << Bug( "%p %s without Track Segment"
, autoSegment
, getString(autoSegment).c_str() ) << endl;
} else
trackSegment->_check();
}
}
return coherency;
}
void KiteEngine::finalizeLayout ()
{
ltrace(90) << "KiteEngine::finalizeLayout()" << endl;
if (getState() > Katabatic::EngineDriving) return;
ltracein(90);
setState( Katabatic::EngineDriving );
_gutKite();
KatabaticEngine::finalizeLayout();
ltrace(90) << "State: " << getState() << endl;
ltraceout(90);
}
void KiteEngine::_gutKite ()
{
ltrace(90) << "KiteEngine::_gutKite()" << endl;
ltracein(90);
ltrace(90) << "State: " << getState() << endl;
if (getState() < Katabatic::EngineGutted) {
Session::open( this );
size_t maxDepth = getRoutingGauge()->getDepth();
for ( size_t depth=0 ; depth < maxDepth ; depth++ ) {
_routingPlanes[depth]->destroy();
}
Session::close();
}
ltraceout(90);
}
TrackElement* KiteEngine::_lookup ( Segment* segment ) const
{
AutoSegment* autoSegment = KatabaticEngine::_lookup( segment );
if (not autoSegment or not autoSegment->isCanonical()) return NULL;
return _lookup( autoSegment );
}
void KiteEngine::_check ( Net* net ) const
{
cerr << " o Checking " << net << endl;
forEach ( Segment*, isegment, net->getComponents().getSubSet<Segment*>() ) {
TrackElement* trackSegment = _lookup( *isegment );
if (trackSegment) {
trackSegment->_check();
AutoContact* autoContact = trackSegment->base()->getAutoSource();
if (autoContact) autoContact->checkTopology ();
autoContact = trackSegment->base()->getAutoTarget();
if (autoContact) autoContact->checkTopology ();
}
}
}
string KiteEngine::_getTypeName () const
{ return "Kite::KiteEngine"; }
string KiteEngine::_getString () const
{
ostringstream os;
os << "<" << "KiteEngine " << _cell->getName () << ">";
return os.str();
}
Record* KiteEngine::_getRecord () const
{
Record* record = KatabaticEngine::_getRecord ();
if (record) {
record->add( getSlot( "_routingPlanes", &_routingPlanes ) );
record->add( getSlot( "_configuration", _configuration ) );
* ./Kite: - 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.
2010-12-30 12:42:17 -06:00
}
return record;
}
} // Kite namespace.