riscv
/
coriolis
mirror of https://gitlab.lip6.fr/vlsi-eda/coriolis.git
1
0
Fork 0
Code Issues Projects Releases Wiki Activity
coriolis/kite/src/NegociateWindow.cpp

604 lines
19 KiB
C++
Raw Normal View History

* ./hurricane/src/hviewer, ./coriolis/src/crlcore, ./coriolis/src/knik, ./coriolis/src/katabatic, ./coriolis/src/kite, ./coriolis/src/equinox, ./coriolis/src/solstice, ./coriolis/src/ispd: - SVN MOVE: Source tree simplification & uniformisation. Now all tools are at the same level, directly under the root of the repository. No more "coriolis/src".
2010-03-09 09:24:55 -06:00
// -*- C++ -*-
//
// This file is part of the Coriolis Software.
* All Tools: - New: Added FreeBSD/Ubuntu patches from Otacilio De Araujo (<otaciliodearaujo@gmail.com>). * ./kite: - New: Makes use of cbug to display the (scary) debug messages. - Change: Re-enable deletion of Knik in Kite destructor.
2012-12-03 02:29:28 -06:00
// Copyright (c) UPMC/LIP6 2008-2012, All Rights Reserved
* ./hurricane/src/hviewer, ./coriolis/src/crlcore, ./coriolis/src/knik, ./coriolis/src/katabatic, ./coriolis/src/kite, ./coriolis/src/equinox, ./coriolis/src/solstice, ./coriolis/src/ispd: - SVN MOVE: Source tree simplification & uniformisation. Now all tools are at the same level, directly under the root of the repository. No more "coriolis/src".
2010-03-09 09:24:55 -06:00
//
* ./kite: - New: In RoutingEvent, while routing a full chip, transient full blockages may happens due to the initial position of some perpandiculars on theirs optimal positions. Check for it and perform a "ripupPerpandicular" with the perpandiculars re-routed first (on normal operation, this is the reverse). - Change: In NegociateWindow::NegociateOverlapCost(), do not account terminals if the layer is *above* METAL3 as it is unlikely that it's truly directly connected to a terminal (true at least for our designs). - Change: In RoutingEvent::State, add a "fullBlocked" flag to perform the full blockage direction while building the state object. - Change: In RoutingEvent::cacheAxisHint(), when the TrackSegment has parent, uses the parent current axis position and not it's axis hint. - Change: In Manipulator::insertInTrack(), when the overlapped segment is a Local and is completly enclosed (shrinkLeft & shrinkRight), no longer rip it up but force a shrink left/right instead. - Bug: In RoutingEvent, the event sorting function was sorting in the *wrong* order! The less prioritary first! With the correct sort, we won an additionnal 30% in speed (total: 69%). From the "reference" time we have a 3.2 speed-up. And we can successfully process denser designs...
2011-01-09 12:08:57 -06:00
// +-----------------------------------------------------------------+
* ./hurricane/src/hviewer, ./coriolis/src/crlcore, ./coriolis/src/knik, ./coriolis/src/katabatic, ./coriolis/src/kite, ./coriolis/src/equinox, ./coriolis/src/solstice, ./coriolis/src/ispd: - SVN MOVE: Source tree simplification & uniformisation. Now all tools are at the same level, directly under the root of the repository. No more "coriolis/src".
2010-03-09 09:24:55 -06:00
// | C O R I O L I S |
// | K i t e - D e t a i l e d R o u t e r |
// | |
// | Author : Jean-Paul CHAPUT |
// | E-mail : Jean-Paul.Chaput@asim.lip6.fr |
// | =============================================================== |
// | C++ Module : "./NegociateWindow.cpp" |
* ./kite: - New: In RoutingEvent, while routing a full chip, transient full blockages may happens due to the initial position of some perpandiculars on theirs optimal positions. Check for it and perform a "ripupPerpandicular" with the perpandiculars re-routed first (on normal operation, this is the reverse). - Change: In NegociateWindow::NegociateOverlapCost(), do not account terminals if the layer is *above* METAL3 as it is unlikely that it's truly directly connected to a terminal (true at least for our designs). - Change: In RoutingEvent::State, add a "fullBlocked" flag to perform the full blockage direction while building the state object. - Change: In RoutingEvent::cacheAxisHint(), when the TrackSegment has parent, uses the parent current axis position and not it's axis hint. - Change: In Manipulator::insertInTrack(), when the overlapped segment is a Local and is completly enclosed (shrinkLeft & shrinkRight), no longer rip it up but force a shrink left/right instead. - Bug: In RoutingEvent, the event sorting function was sorting in the *wrong* order! The less prioritary first! With the correct sort, we won an additionnal 30% in speed (total: 69%). From the "reference" time we have a 3.2 speed-up. And we can successfully process denser designs...
2011-01-09 12:08:57 -06:00
// +-----------------------------------------------------------------+
* ./hurricane/src/hviewer, ./coriolis/src/crlcore, ./coriolis/src/knik, ./coriolis/src/katabatic, ./coriolis/src/kite, ./coriolis/src/equinox, ./coriolis/src/solstice, ./coriolis/src/ispd: - SVN MOVE: Source tree simplification & uniformisation. Now all tools are at the same level, directly under the root of the repository. No more "coriolis/src".
2010-03-09 09:24:55 -06:00
* ./kite: - Change: Big cleanup. Supress all the GCell level: GCell, GCellGrid, GCellRoutingSet. Blockages now managed inside the BuildPower. - New: In NegociateWindow, Histogram of GCell densities in Gnuplot format.
2010-12-12 15:42:57 -06:00
#include <vector>
* ./hurricane/src/hviewer, ./coriolis/src/crlcore, ./coriolis/src/knik, ./coriolis/src/katabatic, ./coriolis/src/kite, ./coriolis/src/equinox, ./coriolis/src/solstice, ./coriolis/src/ispd: - SVN MOVE: Source tree simplification & uniformisation. Now all tools are at the same level, directly under the root of the repository. No more "coriolis/src".
2010-03-09 09:24:55 -06:00
#include <algorithm>
#include <iomanip>
* ./kite: - Change: Big cleanup. Supress all the GCell level: GCell, GCellGrid, GCellRoutingSet. Blockages now managed inside the BuildPower. - New: In NegociateWindow, Histogram of GCell densities in Gnuplot format.
2010-12-12 15:42:57 -06:00
#include "hurricane/Warning.h"
* ./hurricane/src/hviewer, ./coriolis/src/crlcore, ./coriolis/src/knik, ./coriolis/src/katabatic, ./coriolis/src/kite, ./coriolis/src/equinox, ./coriolis/src/solstice, ./coriolis/src/ispd: - SVN MOVE: Source tree simplification & uniformisation. Now all tools are at the same level, directly under the root of the repository. No more "coriolis/src".
2010-03-09 09:24:55 -06:00
#include "hurricane/Bug.h"
#include "hurricane/RoutingPad.h"
#include "hurricane/Net.h"
#include "hurricane/Cell.h"
* Most of tools: - Bug: In top CMakeLists.txt the SETUP_PROJECT_PATHS was not inserting X_USER_TOP *before* X_TOP, thus potentially allowing an obsolete system-wide configuration to shadow an up-to-date local one. * ./kite: - Change: In Configuration/KiteEngine, propagate changes in Configuration in Kite. Note that a *second* singleton is created at Kite level. - New: KiteEngine::printConfiguration() method to allow quick checks of the router's main parameters. - New: ConfigurationWidget to display/sets all the revelant Katabatic/ Kite configuration parameters. To be used with the TabSettings from the ControllerWidget. The ConfigurationWidget is bound to the default (static) Kite configuration, as once it has been run you obviously cannot change the router's settings.
2010-04-28 10:44:07 -05:00
#include "crlcore/Utilities.h"
* ./hurricane/src/hviewer, ./coriolis/src/crlcore, ./coriolis/src/knik, ./coriolis/src/katabatic, ./coriolis/src/kite, ./coriolis/src/equinox, ./coriolis/src/solstice, ./coriolis/src/ispd: - SVN MOVE: Source tree simplification & uniformisation. Now all tools are at the same level, directly under the root of the repository. No more "coriolis/src".
2010-03-09 09:24:55 -06:00
#include "crlcore/AllianceFramework.h"
* ./kite: - New: In Configuration & RoutingEvent, adds support for minimal break length by layer (METAL1 through METAL7). - New: Added measures for final detailed wirelength, wirelength expansion ratio, failed wirelength, number of processeds events and number of unique events.
2010-08-22 07:38:27 -05:00
#include "crlcore/Measures.h"
* ./kite: - Change: Removed Histogram, moved into crlcore. - Change: In GraphicKiteEngine, add a menu entry to dump the measurments.
2010-12-15 09:12:00 -06:00
#include "crlcore/Histogram.h"
* ./kite: - Change: Big cleanup. Supress all the GCell level: GCell, GCellGrid, GCellRoutingSet. Blockages now managed inside the BuildPower. - New: In NegociateWindow, Histogram of GCell densities in Gnuplot format.
2010-12-12 15:42:57 -06:00
#include "katabatic/AutoContact.h"
#include "katabatic/GCellGrid.h"
* ./hurricane/src/hviewer, ./coriolis/src/crlcore, ./coriolis/src/knik, ./coriolis/src/katabatic, ./coriolis/src/kite, ./coriolis/src/equinox, ./coriolis/src/solstice, ./coriolis/src/ispd: - SVN MOVE: Source tree simplification & uniformisation. Now all tools are at the same level, directly under the root of the repository. No more "coriolis/src".
2010-03-09 09:24:55 -06:00
#include "kite/DataNegociate.h"
#include "kite/TrackElement.h"
#include "kite/TrackMarker.h"
#include "kite/TrackCost.h"
#include "kite/Track.h"
* ./kite: - Change: Big cleanup. Supress all the GCell level: GCell, GCellGrid, GCellRoutingSet. Blockages now managed inside the BuildPower. - New: In NegociateWindow, Histogram of GCell densities in Gnuplot format.
2010-12-12 15:42:57 -06:00
#include "kite/TrackSegment.h"
* ./hurricane/src/hviewer, ./coriolis/src/crlcore, ./coriolis/src/knik, ./coriolis/src/katabatic, ./coriolis/src/kite, ./coriolis/src/equinox, ./coriolis/src/solstice, ./coriolis/src/ispd: - SVN MOVE: Source tree simplification & uniformisation. Now all tools are at the same level, directly under the root of the repository. No more "coriolis/src".
2010-03-09 09:24:55 -06:00
#include "kite/RoutingPlane.h"
#include "kite/RoutingEventQueue.h"
#include "kite/RoutingEventHistory.h"
* ./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
#include "kite/RoutingEventLoop.h"
* ./hurricane/src/hviewer, ./coriolis/src/crlcore, ./coriolis/src/knik, ./coriolis/src/katabatic, ./coriolis/src/kite, ./coriolis/src/equinox, ./coriolis/src/solstice, ./coriolis/src/ispd: - SVN MOVE: Source tree simplification & uniformisation. Now all tools are at the same level, directly under the root of the repository. No more "coriolis/src".
2010-03-09 09:24:55 -06:00
#include "kite/NegociateWindow.h"
#include "kite/KiteEngine.h"
namespace {
using namespace std;
using namespace Hurricane;
using namespace CRL;
using namespace Kite;
void NegociateOverlapCost ( const TrackElement* segment, TrackCost& cost )
{
Interval intersect = segment->getCanonicalInterval();
* ./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
if ( not intersect.intersect(cost.getInterval()) ) return;
* ./hurricane/src/hviewer, ./coriolis/src/crlcore, ./coriolis/src/knik, ./coriolis/src/katabatic, ./coriolis/src/kite, ./coriolis/src/equinox, ./coriolis/src/solstice, ./coriolis/src/ispd: - SVN MOVE: Source tree simplification & uniformisation. Now all tools are at the same level, directly under the root of the repository. No more "coriolis/src".
2010-03-09 09:24:55 -06:00
* ./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 ( segment->isBlockage() or segment->isFixed() ) {
* ./kite: - Change: Big cleanup. Supress all the GCell level: GCell, GCellGrid, GCellRoutingSet. Blockages now managed inside the BuildPower. - New: In NegociateWindow, Histogram of GCell densities in Gnuplot format.
2010-12-12 15:42:57 -06:00
ltrace(200) << "Infinite cost from: " << segment << endl;
* ./hurricane/src/hviewer, ./coriolis/src/crlcore, ./coriolis/src/knik, ./coriolis/src/katabatic, ./coriolis/src/kite, ./coriolis/src/equinox, ./coriolis/src/solstice, ./coriolis/src/ispd: - SVN MOVE: Source tree simplification & uniformisation. Now all tools are at the same level, directly under the root of the repository. No more "coriolis/src".
2010-03-09 09:24:55 -06:00
cost.setInfinite ();
cost.setOverlap ();
cost.setHardOverlap ();
cost.setBlockage ();
return;
}
if ( cost.getInterval().getVMax() > intersect.getVMax() ) cost.setLeftOverlap();
if ( cost.getInterval().getVMin() < intersect.getVMin() ) cost.setRightOverlap();
* ./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
//cost.setLonguestOverlap ( intersect.getSize() );
//intersect.intersection ( cost.getInterval() );
* ./hurricane/src/hviewer, ./coriolis/src/crlcore, ./coriolis/src/knik, ./coriolis/src/katabatic, ./coriolis/src/kite, ./coriolis/src/equinox, ./coriolis/src/solstice, ./coriolis/src/ispd: - SVN MOVE: Source tree simplification & uniformisation. Now all tools are at the same level, directly under the root of the repository. No more "coriolis/src".
2010-03-09 09:24:55 -06:00
if ( not intersect.contains(cost.getInterval()) )
intersect.intersection ( cost.getInterval() );
* ./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
else {
cost.setLonguestOverlap ( intersect.getSize() );
cost.setGlobalEnclosed ();
}
* ./hurricane/src/hviewer, ./coriolis/src/crlcore, ./coriolis/src/knik, ./coriolis/src/katabatic, ./coriolis/src/kite, ./coriolis/src/equinox, ./coriolis/src/solstice, ./coriolis/src/ispd: - SVN MOVE: Source tree simplification & uniformisation. Now all tools are at the same level, directly under the root of the repository. No more "coriolis/src".
2010-03-09 09:24:55 -06:00
DataNegociate* data = segment->getDataNegociate ();
if ( not data ) return;
* ./kite: - Change: Big cleanup. Supress all the GCell level: GCell, GCellGrid, GCellRoutingSet. Blockages now managed inside the BuildPower. - New: In NegociateWindow, Histogram of GCell densities in Gnuplot format.
2010-12-12 15:42:57 -06:00
cost.mergeRipupCount ( data->getRipupCount() );
if ( segment->isLocal() ) {
cost.mergeDataState ( data->getState() );
if ( data->getState() >= DataNegociate::LocalVsGlobal ) {
ltrace(200) << "MaximumSlack/LocalVsGlobal for " << segment << endl;
* ./hurricane/src/hviewer, ./coriolis/src/crlcore, ./coriolis/src/knik, ./coriolis/src/katabatic, ./coriolis/src/kite, ./coriolis/src/equinox, ./coriolis/src/solstice, ./coriolis/src/ispd: - SVN MOVE: Source tree simplification & uniformisation. Now all tools are at the same level, directly under the root of the repository. No more "coriolis/src".
2010-03-09 09:24:55 -06:00
}
}
* ./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
if ( segment->isGlobal() ) {
//if ( data->getState() >= DataNegociate::ConflictSolve1 ) {
cost.setOverlapGlobal();
//}
* ./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
if ( (cost.getFlags() & TrackCost::LocalAndTopDepth)
and (data->getState() >= DataNegociate::MoveUp) ) {
cost.setInfinite ();
cost.setOverlap ();
cost.setHardOverlap ();
return;
}
* ./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
}
* ./kite: - Change: Big cleanup. Supress all the GCell level: GCell, GCellGrid, GCellRoutingSet. Blockages now managed inside the BuildPower. - New: In NegociateWindow, Histogram of GCell densities in Gnuplot format.
2010-12-12 15:42:57 -06:00
// if ( data->getRipupCount() > 3 ) {
// ltrace(200) << "Infinite cost from: " << segment << endl;
// cost.setFixed ();
// cost.setInfinite ();
// cost.setOverlap ();
// cost.setHardOverlap ();
// return;
// }
* ./hurricane/src/hviewer, ./coriolis/src/crlcore, ./coriolis/src/knik, ./coriolis/src/katabatic, ./coriolis/src/kite, ./coriolis/src/equinox, ./coriolis/src/solstice, ./coriolis/src/ispd: - SVN MOVE: Source tree simplification & uniformisation. Now all tools are at the same level, directly under the root of the repository. No more "coriolis/src".
2010-03-09 09:24:55 -06:00
cost.setOverlap ();
* ./kite: - New: In RoutingEvent, while routing a full chip, transient full blockages may happens due to the initial position of some perpandiculars on theirs optimal positions. Check for it and perform a "ripupPerpandicular" with the perpandiculars re-routed first (on normal operation, this is the reverse). - Change: In NegociateWindow::NegociateOverlapCost(), do not account terminals if the layer is *above* METAL3 as it is unlikely that it's truly directly connected to a terminal (true at least for our designs). - Change: In RoutingEvent::State, add a "fullBlocked" flag to perform the full blockage direction while building the state object. - Change: In RoutingEvent::cacheAxisHint(), when the TrackSegment has parent, uses the parent current axis position and not it's axis hint. - Change: In Manipulator::insertInTrack(), when the overlapped segment is a Local and is completly enclosed (shrinkLeft & shrinkRight), no longer rip it up but force a shrink left/right instead. - Bug: In RoutingEvent, the event sorting function was sorting in the *wrong* order! The less prioritary first! With the correct sort, we won an additionnal 30% in speed (total: 69%). From the "reference" time we have a 3.2 speed-up. And we can successfully process denser designs...
2011-01-09 12:08:57 -06:00
if ( segment->isLocal() or (Session::getRoutingGauge()->getLayerDepth(segment->getLayer()) < 3) )
cost.incTerminals ( data->getCost().getTerminals()*100 );
* ./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
ltrace(200) << "| Increment Delta: " << DbU::getValueString(intersect.getSize()) << endl;
cost.incDelta ( intersect.getSize() );
* ./hurricane/src/hviewer, ./coriolis/src/crlcore, ./coriolis/src/knik, ./coriolis/src/katabatic, ./coriolis/src/kite, ./coriolis/src/equinox, ./coriolis/src/solstice, ./coriolis/src/ispd: - SVN MOVE: Source tree simplification & uniformisation. Now all tools are at the same level, directly under the root of the repository. No more "coriolis/src".
2010-03-09 09:24:55 -06:00
}
void loadRoutingPads ( NegociateWindow* nw )
{
AllianceFramework* af = AllianceFramework::get ();
RoutingGauge* rg = nw->getKiteEngine()->getRoutingGauge();
forEach ( Net*, inet, nw->getCell()->getNets() ) {
if ( inet->getType() == Net::Type::POWER ) continue;
if ( inet->getType() == Net::Type::GROUND ) continue;
if ( inet->getType() == Net::Type::CLOCK ) 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 ( af->isBLOCKAGE(inet->getName()) ) continue;
* ./hurricane/src/hviewer, ./coriolis/src/crlcore, ./coriolis/src/knik, ./coriolis/src/katabatic, ./coriolis/src/kite, ./coriolis/src/equinox, ./coriolis/src/solstice, ./coriolis/src/ispd: - SVN MOVE: Source tree simplification & uniformisation. Now all tools are at the same level, directly under the root of the repository. No more "coriolis/src".
2010-03-09 09:24:55 -06:00
forEach ( RoutingPad*, irp, inet->getRoutingPads() ) {
size_t depth = rg->getLayerDepth(irp->getLayer());
if ( depth > 0 ) continue;
if ( depth == 0 ) {
TrackMarker::create ( *irp, 1 );
* ./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
//TrackMarker::create ( *irp, 2 );
* ./hurricane/src/hviewer, ./coriolis/src/crlcore, ./coriolis/src/knik, ./coriolis/src/katabatic, ./coriolis/src/kite, ./coriolis/src/equinox, ./coriolis/src/solstice, ./coriolis/src/ispd: - SVN MOVE: Source tree simplification & uniformisation. Now all tools are at the same level, directly under the root of the repository. No more "coriolis/src".
2010-03-09 09:24:55 -06:00
}
}
}
}
} // End of local namespace.
namespace Kite {
using std::cerr;
using std::endl;
using std::setw;
using std::left;
using std::right;
using std::setprecision;
* ./kite: - Change: Big cleanup. Supress all the GCell level: GCell, GCellGrid, GCellRoutingSet. Blockages now managed inside the BuildPower. - New: In NegociateWindow, Histogram of GCell densities in Gnuplot format.
2010-12-12 15:42:57 -06:00
using Hurricane::Warning;
* ./hurricane/src/hviewer, ./coriolis/src/crlcore, ./coriolis/src/knik, ./coriolis/src/katabatic, ./coriolis/src/kite, ./coriolis/src/equinox, ./coriolis/src/solstice, ./coriolis/src/ispd: - SVN MOVE: Source tree simplification & uniformisation. Now all tools are at the same level, directly under the root of the repository. No more "coriolis/src".
2010-03-09 09:24:55 -06:00
using Hurricane::Bug;
using Hurricane::tab;
using Hurricane::inltrace;
using Hurricane::ltracein;
using Hurricane::ltraceout;
using Hurricane::ForEachIterator;
* ./kite: - Change: Removed Histogram, moved into crlcore. - Change: In GraphicKiteEngine, add a menu entry to dump the measurments.
2010-12-15 09:12:00 -06:00
using CRL::Histogram;
* ./kite: - New: In Configuration & RoutingEvent, adds support for minimal break length by layer (METAL1 through METAL7). - New: Added measures for final detailed wirelength, wirelength expansion ratio, failed wirelength, number of processeds events and number of unique events.
2010-08-22 07:38:27 -05:00
using CRL::addMeasure;
* ./kite: - Change: Big cleanup. Supress all the GCell level: GCell, GCellGrid, GCellRoutingSet. Blockages now managed inside the BuildPower. - New: In NegociateWindow, Histogram of GCell densities in Gnuplot format.
2010-12-12 15:42:57 -06:00
using Katabatic::AutoContact;
* ./hurricane/src/hviewer, ./coriolis/src/crlcore, ./coriolis/src/knik, ./coriolis/src/katabatic, ./coriolis/src/kite, ./coriolis/src/equinox, ./coriolis/src/solstice, ./coriolis/src/ispd: - SVN MOVE: Source tree simplification & uniformisation. Now all tools are at the same level, directly under the root of the repository. No more "coriolis/src".
2010-03-09 09:24:55 -06:00
// -------------------------------------------------------------------
// Class : "NegociateWindow".
* ./kite: - Change: Big cleanup. Supress all the GCell level: GCell, GCellGrid, GCellRoutingSet. Blockages now managed inside the BuildPower. - New: In NegociateWindow, Histogram of GCell densities in Gnuplot format.
2010-12-12 15:42:57 -06:00
NegociateWindow::NegociateWindow ( KiteEngine* kite )
: _slowMotion (0)
, _interrupt (false)
, _kite (kite)
, _gcells ()
, _segments ()
, _eventQueue ()
, _eventHistory()
* ./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
, _eventLoop (10,50)
* ./kite: - Change: Big cleanup. Supress all the GCell level: GCell, GCellGrid, GCellRoutingSet. Blockages now managed inside the BuildPower. - New: In NegociateWindow, Histogram of GCell densities in Gnuplot format.
2010-12-12 15:42:57 -06:00
{ }
* ./hurricane/src/hviewer, ./coriolis/src/crlcore, ./coriolis/src/knik, ./coriolis/src/katabatic, ./coriolis/src/kite, ./coriolis/src/equinox, ./coriolis/src/solstice, ./coriolis/src/ispd: - SVN MOVE: Source tree simplification & uniformisation. Now all tools are at the same level, directly under the root of the repository. No more "coriolis/src".
2010-03-09 09:24:55 -06:00
* ./kite: - Change: Big cleanup. Supress all the GCell level: GCell, GCellGrid, GCellRoutingSet. Blockages now managed inside the BuildPower. - New: In NegociateWindow, Histogram of GCell densities in Gnuplot format.
2010-12-12 15:42:57 -06:00
NegociateWindow* NegociateWindow::create ( KiteEngine* kite )
* ./hurricane/src/hviewer, ./coriolis/src/crlcore, ./coriolis/src/knik, ./coriolis/src/katabatic, ./coriolis/src/kite, ./coriolis/src/equinox, ./coriolis/src/solstice, ./coriolis/src/ispd: - SVN MOVE: Source tree simplification & uniformisation. Now all tools are at the same level, directly under the root of the repository. No more "coriolis/src".
2010-03-09 09:24:55 -06:00
{
* ./kite: - Change: Big cleanup. Supress all the GCell level: GCell, GCellGrid, GCellRoutingSet. Blockages now managed inside the BuildPower. - New: In NegociateWindow, Histogram of GCell densities in Gnuplot format.
2010-12-12 15:42:57 -06:00
NegociateWindow* negociateWindow = new NegociateWindow ( kite );
* ./hurricane/src/hviewer, ./coriolis/src/crlcore, ./coriolis/src/knik, ./coriolis/src/katabatic, ./coriolis/src/kite, ./coriolis/src/equinox, ./coriolis/src/solstice, ./coriolis/src/ispd: - SVN MOVE: Source tree simplification & uniformisation. Now all tools are at the same level, directly under the root of the repository. No more "coriolis/src".
2010-03-09 09:24:55 -06:00
return negociateWindow;
}
NegociateWindow::~NegociateWindow ()
* ./kite: - Change: Big cleanup. Supress all the GCell level: GCell, GCellGrid, GCellRoutingSet. Blockages now managed inside the BuildPower. - New: In NegociateWindow, Histogram of GCell densities in Gnuplot format.
2010-12-12 15:42:57 -06:00
{ }
* ./hurricane/src/hviewer, ./coriolis/src/crlcore, ./coriolis/src/knik, ./coriolis/src/katabatic, ./coriolis/src/kite, ./coriolis/src/equinox, ./coriolis/src/solstice, ./coriolis/src/ispd: - SVN MOVE: Source tree simplification & uniformisation. Now all tools are at the same level, directly under the root of the repository. No more "coriolis/src".
2010-03-09 09:24:55 -06:00
void NegociateWindow::destroy ()
{ delete this; }
Cell* NegociateWindow::getCell () const
{ return _kite->getCell(); }
* ./kite: - Change: Big cleanup. Supress all the GCell level: GCell, GCellGrid, GCellRoutingSet. Blockages now managed inside the BuildPower. - New: In NegociateWindow, Histogram of GCell densities in Gnuplot format.
2010-12-12 15:42:57 -06:00
void NegociateWindow::setGCells ( const Katabatic::GCellVector& gcells )
* ./hurricane/src/hviewer, ./coriolis/src/crlcore, ./coriolis/src/knik, ./coriolis/src/katabatic, ./coriolis/src/kite, ./coriolis/src/equinox, ./coriolis/src/solstice, ./coriolis/src/ispd: - SVN MOVE: Source tree simplification & uniformisation. Now all tools are at the same level, directly under the root of the repository. No more "coriolis/src".
2010-03-09 09:24:55 -06:00
{
* ./kite: - Change: Big cleanup. Supress all the GCell level: GCell, GCellGrid, GCellRoutingSet. Blockages now managed inside the BuildPower. - New: In NegociateWindow, Histogram of GCell densities in Gnuplot format.
2010-12-12 15:42:57 -06:00
_gcells = gcells;
//sort ( _gcells.begin(), _gcells.end(), Katabatic::GCell::CompareGCellById() );
* ./hurricane/src/hviewer, ./coriolis/src/crlcore, ./coriolis/src/knik, ./coriolis/src/katabatic, ./coriolis/src/kite, ./coriolis/src/equinox, ./coriolis/src/solstice, ./coriolis/src/ispd: - SVN MOVE: Source tree simplification & uniformisation. Now all tools are at the same level, directly under the root of the repository. No more "coriolis/src".
2010-03-09 09:24:55 -06:00
loadRoutingPads ( this );
Session::revalidate ();
TrackElement* segment;
TrackElementLut lut = Session::getKiteEngine()->_getTrackElementLut();
TrackElementLut::iterator it = lut.begin ();
for ( ; it != lut.end() ; it++ ) {
segment = it->second;
segment->getDataNegociate()->update();
}
* ./kite: - Change: Big cleanup. Supress all the GCell level: GCell, GCellGrid, GCellRoutingSet. Blockages now managed inside the BuildPower. - New: In NegociateWindow, Histogram of GCell densities in Gnuplot format.
2010-12-12 15:42:57 -06:00
_statistics.setGCellsCount ( _gcells.size() );
* ./hurricane/src/hviewer, ./coriolis/src/crlcore, ./coriolis/src/knik, ./coriolis/src/katabatic, ./coriolis/src/kite, ./coriolis/src/equinox, ./coriolis/src/solstice, ./coriolis/src/ispd: - SVN MOVE: Source tree simplification & uniformisation. Now all tools are at the same level, directly under the root of the repository. No more "coriolis/src".
2010-03-09 09:24:55 -06:00
}
void NegociateWindow::addInsertEvent ( TrackElement* segment, unsigned int level )
{
DataNegociate* data = segment->getDataNegociate();
if ( not data or not data->hasRoutingEvent() )
_eventQueue.add ( segment, level );
else
cerr << Bug("NegociateWidow::addInsertEvent(): Try to adds twice the same TrackElement event."
"\n %p:%s."
,(void*)segment->base()->base()
,getString(segment).c_str()
) << endl;
}
* ./kite: - Change: Big cleanup. Supress all the GCell level: GCell, GCellGrid, GCellRoutingSet. Blockages now managed inside the BuildPower. - New: In NegociateWindow, Histogram of GCell densities in Gnuplot format.
2010-12-12 15:42:57 -06:00
TrackElement* NegociateWindow::addTrackSegment ( AutoSegment* autoSegment, bool loading )
* ./hurricane/src/hviewer, ./coriolis/src/crlcore, ./coriolis/src/knik, ./coriolis/src/katabatic, ./coriolis/src/kite, ./coriolis/src/equinox, ./coriolis/src/solstice, ./coriolis/src/ispd: - SVN MOVE: Source tree simplification & uniformisation. Now all tools are at the same level, directly under the root of the repository. No more "coriolis/src".
2010-03-09 09:24:55 -06:00
{
* ./kite: - Change: Big cleanup. Supress all the GCell level: GCell, GCellGrid, GCellRoutingSet. Blockages now managed inside the BuildPower. - New: In NegociateWindow, Histogram of GCell densities in Gnuplot format.
2010-12-12 15:42:57 -06:00
ltrace(200) << "NegociateWindow::addTrackSegment() - " << autoSegment << endl;
ltracein(159);
// Special case: fixed AutoSegments must not interfere with blockages.
// Ugly: uses of getExtensionCap().
if ( autoSegment->isFixed() ) {
RoutingPlane* plane = Session::getKiteEngine()->getRoutingPlaneByLayer(autoSegment->getLayer());
Track* track = plane->getTrackByPosition ( autoSegment->getAxis() );
size_t begin;
size_t end;
Interval fixedSpan;
Interval blockageSpan;
autoSegment->getCanonical ( fixedSpan );
fixedSpan.inflate ( Session::getExtensionCap()-1 );
track->getOverlapBounds ( fixedSpan, begin, end );
for ( ; (begin < end) ; begin++ ) {
TrackElement* other = track->getSegment(begin);
ltrace(200) << "| overlap: " << other << endl;
if ( not other->isBlockage() ) continue;
other->getCanonical ( blockageSpan );
blockageSpan.inflate(Session::getExtensionCap());
ltrace(200) << " fixed:" << fixedSpan << " vs. blockage:" << blockageSpan << endl;
if ( not fixedSpan.intersect(blockageSpan) ) continue;
// Overlap between fixed & blockage.
ltrace(200) << "* Blockage overlap: " << autoSegment << endl;
Session::destroyRequest ( autoSegment );
cerr << Warning("Overlap between fixed %s and blockage at %s."
,getString(autoSegment).c_str(),getString(blockageSpan).c_str()) << endl;
return NULL;
}
* ./hurricane/src/hviewer, ./coriolis/src/crlcore, ./coriolis/src/knik, ./coriolis/src/katabatic, ./coriolis/src/kite, ./coriolis/src/equinox, ./coriolis/src/solstice, ./coriolis/src/ispd: - SVN MOVE: Source tree simplification & uniformisation. Now all tools are at the same level, directly under the root of the repository. No more "coriolis/src".
2010-03-09 09:24:55 -06:00
}
* ./kite: - Change: Big cleanup. Supress all the GCell level: GCell, GCellGrid, GCellRoutingSet. Blockages now managed inside the BuildPower. - New: In NegociateWindow, Histogram of GCell densities in Gnuplot format.
2010-12-12 15:42:57 -06:00
Interval span;
autoSegment = autoSegment->getCanonical ( span );
bool created;
TrackElement* trackSegment = TrackSegment::create ( autoSegment, NULL, created );
if ( not loading )
ltrace(159) << "* lookup: " << autoSegment << endl;
if ( created ) {
ltrace(159) << "* " << trackSegment << endl;
RoutingPlane* plane = Session::getKiteEngine()->getRoutingPlaneByLayer(autoSegment->getLayer());
Track* track = plane->getTrackByPosition ( autoSegment->getAxis() );
Interval uside = autoSegment->getAutoSource()->getGCell()->getUSide ( Constant::perpandicular(autoSegment->getDirection())/*, false */);
if ( track->getAxis() > uside.getVMax() ) track = track->getPrevious();
if ( track->getAxis() < uside.getVMin() ) track = track->getNext();
trackSegment->setAxis ( track->getAxis(), Katabatic::Realignate|Katabatic::AxisSet );
trackSegment->invalidate ();
if ( trackSegment->isFixed() ) {
Session::addInsertEvent ( trackSegment, track );
} else {
_segments.push_back ( trackSegment );
}
}
if ( not created and not loading ) {
ltrace(200) << "TrackSegment already exists (and not in loading stage)." << endl;
}
ltraceout(159);
return trackSegment;
* ./hurricane/src/hviewer, ./coriolis/src/crlcore, ./coriolis/src/knik, ./coriolis/src/katabatic, ./coriolis/src/kite, ./coriolis/src/equinox, ./coriolis/src/solstice, ./coriolis/src/ispd: - SVN MOVE: Source tree simplification & uniformisation. Now all tools are at the same level, directly under the root of the repository. No more "coriolis/src".
2010-03-09 09:24:55 -06:00
}
* ./kite: - Change: Big cleanup. Supress all the GCell level: GCell, GCellGrid, GCellRoutingSet. Blockages now managed inside the BuildPower. - New: In NegociateWindow, Histogram of GCell densities in Gnuplot format.
2010-12-12 15:42:57 -06:00
double NegociateWindow::computeWirelength ()
* ./hurricane/src/hviewer, ./coriolis/src/crlcore, ./coriolis/src/knik, ./coriolis/src/katabatic, ./coriolis/src/kite, ./coriolis/src/equinox, ./coriolis/src/solstice, ./coriolis/src/ispd: - SVN MOVE: Source tree simplification & uniformisation. Now all tools are at the same level, directly under the root of the repository. No more "coriolis/src".
2010-03-09 09:24:55 -06:00
{
* ./kite: - Change: Big cleanup. Supress all the GCell level: GCell, GCellGrid, GCellRoutingSet. Blockages now managed inside the BuildPower. - New: In NegociateWindow, Histogram of GCell densities in Gnuplot format.
2010-12-12 15:42:57 -06:00
set<TrackElement*> accounteds;
double totalWL = 0.0;
for ( size_t igcell=0 ; igcell<_gcells.size() ; ++igcell ) {
double gcellWL = 0.0;
Segment* segment;
TrackElement* trackSegment;
vector<AutoContact*>* contacts = _gcells[igcell]->getContacts();
for ( size_t i=0 ; i<contacts->size() ; i++ ) {
forEach ( Hook*, ihook, (*contacts)[i]->getBodyHook()->getSlaveHooks() ) {
Hook* sourceHook = dynamic_cast<Segment::SourceHook*>(*ihook);
if ( not sourceHook ) continue;
segment = dynamic_cast<Segment*>(sourceHook->getComponent());
trackSegment = Session::lookup ( segment );
if ( trackSegment ) {
if ( accounteds.find(trackSegment) != accounteds.end() ) continue;
accounteds.insert ( trackSegment );
gcellWL += DbU::getLambda ( trackSegment->getLength() );
}
}
* ./hurricane/src/hviewer, ./coriolis/src/crlcore, ./coriolis/src/knik, ./coriolis/src/katabatic, ./coriolis/src/kite, ./coriolis/src/equinox, ./coriolis/src/solstice, ./coriolis/src/ispd: - SVN MOVE: Source tree simplification & uniformisation. Now all tools are at the same level, directly under the root of the repository. No more "coriolis/src".
2010-03-09 09:24:55 -06:00
}
* ./kite: - Change: Big cleanup. Supress all the GCell level: GCell, GCellGrid, GCellRoutingSet. Blockages now managed inside the BuildPower. - New: In NegociateWindow, Histogram of GCell densities in Gnuplot format.
2010-12-12 15:42:57 -06:00
// Partial sum to limit rounding errors.
totalWL += gcellWL;
}
return totalWL;
}
void NegociateWindow::_createRouting ( Katabatic::GCell* gcell )
{
ltrace(200) << "NegociateWindow::_createRouting() - " << gcell << endl;
ltracein(200);
Segment* segment;
AutoSegment* autoSegment;
ltrace(149) << "AutoSegments from AutoContacts" << endl;
vector<AutoContact*>* contacts = gcell->getContacts();
for ( size_t i=0 ; i<contacts->size() ; i++ ) {
forEach ( Component*, component, (*contacts)[i]->getSlaveComponents() ) {
segment = dynamic_cast<Segment*>(*component);
autoSegment = Session::base()->lookup ( segment );
ltrace(149) << autoSegment << endl;
if ( autoSegment and autoSegment->isCanonical() ) {
addTrackSegment ( autoSegment, true );
}
}
* ./hurricane/src/hviewer, ./coriolis/src/crlcore, ./coriolis/src/knik, ./coriolis/src/katabatic, ./coriolis/src/kite, ./coriolis/src/equinox, ./coriolis/src/solstice, ./coriolis/src/ispd: - SVN MOVE: Source tree simplification & uniformisation. Now all tools are at the same level, directly under the root of the repository. No more "coriolis/src".
2010-03-09 09:24:55 -06:00
}
* ./kite: - Change: Big cleanup. Supress all the GCell level: GCell, GCellGrid, GCellRoutingSet. Blockages now managed inside the BuildPower. - New: In NegociateWindow, Histogram of GCell densities in Gnuplot format.
2010-12-12 15:42:57 -06:00
ltrace(149) << "_segments.size():" << _segments.size() << endl;
ltraceout(200);
* ./hurricane/src/hviewer, ./coriolis/src/crlcore, ./coriolis/src/knik, ./coriolis/src/katabatic, ./coriolis/src/kite, ./coriolis/src/equinox, ./coriolis/src/solstice, ./coriolis/src/ispd: - SVN MOVE: Source tree simplification & uniformisation. Now all tools are at the same level, directly under the root of the repository. No more "coriolis/src".
2010-03-09 09:24:55 -06:00
}
* ./kite: - Change: Big cleanup. Supress all the GCell level: GCell, GCellGrid, GCellRoutingSet. Blockages now managed inside the BuildPower. - New: In NegociateWindow, Histogram of GCell densities in Gnuplot format.
2010-12-12 15:42:57 -06:00
size_t NegociateWindow::_negociate ()
* ./hurricane/src/hviewer, ./coriolis/src/crlcore, ./coriolis/src/knik, ./coriolis/src/katabatic, ./coriolis/src/kite, ./coriolis/src/equinox, ./coriolis/src/solstice, ./coriolis/src/ispd: - SVN MOVE: Source tree simplification & uniformisation. Now all tools are at the same level, directly under the root of the repository. No more "coriolis/src".
2010-03-09 09:24:55 -06:00
{
* ./kite: - Change: Big cleanup. Supress all the GCell level: GCell, GCellGrid, GCellRoutingSet. Blockages now managed inside the BuildPower. - New: In NegociateWindow, Histogram of GCell densities in Gnuplot format.
2010-12-12 15:42:57 -06:00
ltrace(150) << "NegociateWindow::_negociate() - " << _segments.size() << endl;
* ./hurricane/src/hviewer, ./coriolis/src/crlcore, ./coriolis/src/knik, ./coriolis/src/katabatic, ./coriolis/src/kite, ./coriolis/src/equinox, ./coriolis/src/solstice, ./coriolis/src/ispd: - SVN MOVE: Source tree simplification & uniformisation. Now all tools are at the same level, directly under the root of the repository. No more "coriolis/src".
2010-03-09 09:24:55 -06:00
ltracein(149);
* ./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
cmess1 << " o Negociation Stage." << endl;
* ./kite: - Change: In TrackFixedSegment, when the source/target coordinates of the fixed segment are outside the grid (i.e. the Cell Abutment Box) truncate them. - Change: In GCell::addTrackSegment, when looking for overlap between the to be inserted TrackSegment and Blockage segment, shrink the Blockage span by one DbU to avoid false overlap when they are just edges connexes. - Bug: In RoutingEvent::slackenTopology, in the local segment FSM, looping on the MaximumSlack state. Now also checks the state count. - Change: In TrackSegment::canMoveUp() & canPivotUp() adds the "reserve" parameter and uses it in RoutingEvent to prevent over saturated GCells. - Change: In Configuration, parameters adjustements: expandStep increased to 0.40. Makes RoutingSets biggers but no noticeable slow down.
2010-04-23 08:14:17 -05:00
unsigned long limit = _kite->getEventsLimit();
* ./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
//unsigned long limit = 25586;
* ./kite: - Change: In TrackFixedSegment, when the source/target coordinates of the fixed segment are outside the grid (i.e. the Cell Abutment Box) truncate them. - Change: In GCell::addTrackSegment, when looking for overlap between the to be inserted TrackSegment and Blockage segment, shrink the Blockage span by one DbU to avoid false overlap when they are just edges connexes. - Bug: In RoutingEvent::slackenTopology, in the local segment FSM, looping on the MaximumSlack state. Now also checks the state count. - Change: In TrackSegment::canMoveUp() & canPivotUp() adds the "reserve" parameter and uses it in RoutingEvent to prevent over saturated GCells. - Change: In Configuration, parameters adjustements: expandStep increased to 0.40. Makes RoutingSets biggers but no noticeable slow down.
2010-04-23 08:14:17 -05:00
* ./hurricane/src/hviewer, ./coriolis/src/crlcore, ./coriolis/src/knik, ./coriolis/src/katabatic, ./coriolis/src/kite, ./coriolis/src/equinox, ./coriolis/src/solstice, ./coriolis/src/ispd: - SVN MOVE: Source tree simplification & uniformisation. Now all tools are at the same level, directly under the root of the repository. No more "coriolis/src".
2010-03-09 09:24:55 -06:00
_eventHistory.clear();
* ./kite: - Change: Big cleanup. Supress all the GCell level: GCell, GCellGrid, GCellRoutingSet. Blockages now managed inside the BuildPower. - New: In NegociateWindow, Histogram of GCell densities in Gnuplot format.
2010-12-12 15:42:57 -06:00
_eventQueue.load ( _segments );
* ./hurricane/src/hviewer, ./coriolis/src/crlcore, ./coriolis/src/knik, ./coriolis/src/katabatic, ./coriolis/src/kite, ./coriolis/src/equinox, ./coriolis/src/solstice, ./coriolis/src/ispd: - SVN MOVE: Source tree simplification & uniformisation. Now all tools are at the same level, directly under the root of the repository. No more "coriolis/src".
2010-03-09 09:24:55 -06:00
size_t count = 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
RoutingEvent::setStage ( RoutingEvent::Negociate );
* ./hurricane/src/hviewer, ./coriolis/src/crlcore, ./coriolis/src/knik, ./coriolis/src/katabatic, ./coriolis/src/kite, ./coriolis/src/equinox, ./coriolis/src/solstice, ./coriolis/src/ispd: - SVN MOVE: Source tree simplification & uniformisation. Now all tools are at the same level, directly under the root of the repository. No more "coriolis/src".
2010-03-09 09:24:55 -06:00
while ( not _eventQueue.empty() and not isInterrupted() ) {
RoutingEvent* event = _eventQueue.pop ();
if (tty::enabled()) {
* ./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
cmess2 << " <event:" << tty::bold << setw(7) << setfill('0')
* ./hurricane/src/hviewer, ./coriolis/src/crlcore, ./coriolis/src/knik, ./coriolis/src/katabatic, ./coriolis/src/kite, ./coriolis/src/equinox, ./coriolis/src/solstice, ./coriolis/src/ispd: - SVN MOVE: Source tree simplification & uniformisation. Now all tools are at the same level, directly under the root of the repository. No more "coriolis/src".
2010-03-09 09:24:55 -06:00
<< RoutingEvent::getProcesseds() << setfill(' ') << tty::reset << ">" << tty::cr;
* ./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
cmess2.flush ();
* ./hurricane/src/hviewer, ./coriolis/src/crlcore, ./coriolis/src/knik, ./coriolis/src/katabatic, ./coriolis/src/kite, ./coriolis/src/equinox, ./coriolis/src/solstice, ./coriolis/src/ispd: - SVN MOVE: Source tree simplification & uniformisation. Now all tools are at the same level, directly under the root of the repository. No more "coriolis/src".
2010-03-09 09:24:55 -06:00
} else {
* ./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
cmess2 << " <event:" << setw(7) << setfill('0')
* ./kite: - New: In RoutingEvent, while routing a full chip, transient full blockages may happens due to the initial position of some perpandiculars on theirs optimal positions. Check for it and perform a "ripupPerpandicular" with the perpandiculars re-routed first (on normal operation, this is the reverse). - Change: In NegociateWindow::NegociateOverlapCost(), do not account terminals if the layer is *above* METAL3 as it is unlikely that it's truly directly connected to a terminal (true at least for our designs). - Change: In RoutingEvent::State, add a "fullBlocked" flag to perform the full blockage direction while building the state object. - Change: In RoutingEvent::cacheAxisHint(), when the TrackSegment has parent, uses the parent current axis position and not it's axis hint. - Change: In Manipulator::insertInTrack(), when the overlapped segment is a Local and is completly enclosed (shrinkLeft & shrinkRight), no longer rip it up but force a shrink left/right instead. - Bug: In RoutingEvent, the event sorting function was sorting in the *wrong* order! The less prioritary first! With the correct sort, we won an additionnal 30% in speed (total: 69%). From the "reference" time we have a 3.2 speed-up. And we can successfully process denser designs...
2011-01-09 12:08:57 -06:00
<< RoutingEvent::getProcesseds() << setfill(' ') << " "
<< event->getEventLevel() << ":" << event->getPriority() << "> "
<< event->getSegment()
//<< "> @" << DbU::getValueString(event->getSegment()->getAxis())
//<< " id:" << event->getSegment()->getId()
//<< " " << event->getSegment()->getNet()->getName()
* ./hurricane/src/hviewer, ./coriolis/src/crlcore, ./coriolis/src/knik, ./coriolis/src/katabatic, ./coriolis/src/kite, ./coriolis/src/equinox, ./coriolis/src/solstice, ./coriolis/src/ispd: - SVN MOVE: Source tree simplification & uniformisation. Now all tools are at the same level, directly under the root of the repository. No more "coriolis/src".
2010-03-09 09:24:55 -06:00
<< endl;
cmess2.flush();
}
* ./kite: - New: In RoutingEvent, while routing a full chip, transient full blockages may happens due to the initial position of some perpandiculars on theirs optimal positions. Check for it and perform a "ripupPerpandicular" with the perpandiculars re-routed first (on normal operation, this is the reverse). - Change: In NegociateWindow::NegociateOverlapCost(), do not account terminals if the layer is *above* METAL3 as it is unlikely that it's truly directly connected to a terminal (true at least for our designs). - Change: In RoutingEvent::State, add a "fullBlocked" flag to perform the full blockage direction while building the state object. - Change: In RoutingEvent::cacheAxisHint(), when the TrackSegment has parent, uses the parent current axis position and not it's axis hint. - Change: In Manipulator::insertInTrack(), when the overlapped segment is a Local and is completly enclosed (shrinkLeft & shrinkRight), no longer rip it up but force a shrink left/right instead. - Bug: In RoutingEvent, the event sorting function was sorting in the *wrong* order! The less prioritary first! With the correct sort, we won an additionnal 30% in speed (total: 69%). From the "reference" time we have a 3.2 speed-up. And we can successfully process denser designs...
2011-01-09 12:08:57 -06:00
event->process ( _eventQueue, _eventHistory, _eventLoop );
* ./hurricane/src/hviewer, ./coriolis/src/crlcore, ./coriolis/src/knik, ./coriolis/src/katabatic, ./coriolis/src/kite, ./coriolis/src/equinox, ./coriolis/src/solstice, ./coriolis/src/ispd: - SVN MOVE: Source tree simplification & uniformisation. Now all tools are at the same level, directly under the root of the repository. No more "coriolis/src".
2010-03-09 09:24:55 -06:00
count++;
* ./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
if ( RoutingEvent::getProcesseds() >= limit ) setInterrupt ( true );
* ./hurricane/src/hviewer, ./coriolis/src/crlcore, ./coriolis/src/knik, ./coriolis/src/katabatic, ./coriolis/src/kite, ./coriolis/src/equinox, ./coriolis/src/solstice, ./coriolis/src/ispd: - SVN MOVE: Source tree simplification & uniformisation. Now all tools are at the same level, directly under the root of the repository. No more "coriolis/src".
2010-03-09 09:24:55 -06:00
}
if (count and tty::enabled()) cmess1 << endl;
* ./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
cmess1 << " o Repair Stage." << endl;
ltrace(200) << "Loadind Repair queue." << endl;
RoutingEvent::setStage ( RoutingEvent::Repair );
* ./kite: - Change: In TrackCost, the ripup cost is now computed from a mix between the number of ripup *and* the data state. Restore a correct balance of the ripping up between segments. - New: In RoutingEvent::State::conflictSolve1(), when a very long Segment enters in conflictSolve1(), still prefers to move it up in one piece, but if it's not possible, it is unlikely that it will overlap with only one or two other Segment. In that cases, blindy break it up in the middle. - New: In RoutingEvent::Manipulator, another overload of makeDogLeg to allow break up at one precise point. - Change/Bug: In RoutingEvent::Manipulator::Relax(), reset the state of the non-overlapping Segment fragments instead of inheriting blindly of the state of the breaked up one. Gives some supplemental slack that allows to converge. - Note: Now the strategy is to disable the uses of GCellRoutingSet by setting the expand step to 0.99. This induces only a slight slow dow and memory increase and passes all the tests... - Mark: This revision passes all the benches except for idct, vld & ieee_division.
2010-05-03 04:16:50 -05:00
for ( size_t i=0 ; (i<_eventHistory.size()) and not isInterrupted() ; i++ ) {
* ./hurricane/src/hviewer, ./coriolis/src/crlcore, ./coriolis/src/knik, ./coriolis/src/katabatic, ./coriolis/src/kite, ./coriolis/src/equinox, ./coriolis/src/solstice, ./coriolis/src/ispd: - SVN MOVE: Source tree simplification & uniformisation. Now all tools are at the same level, directly under the root of the repository. No more "coriolis/src".
2010-03-09 09:24:55 -06:00
RoutingEvent* event = _eventHistory.getNth(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
if ( not event->isCloned() and event->isUnimplemented() ) {
event->reschedule ( _eventQueue, 0 );
* ./hurricane/src/hviewer, ./coriolis/src/crlcore, ./coriolis/src/knik, ./coriolis/src/katabatic, ./coriolis/src/kite, ./coriolis/src/equinox, ./coriolis/src/solstice, ./coriolis/src/ispd: - SVN MOVE: Source tree simplification & uniformisation. Now all tools are at the same level, directly under the root of the repository. No more "coriolis/src".
2010-03-09 09:24:55 -06:00
}
}
* ./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
_eventQueue.commit ();
count = 0;
while ( not _eventQueue.empty() and not isInterrupted() ) {
RoutingEvent* event = _eventQueue.pop ();
if (tty::enabled()) {
cmess2 << " <repair.event:" << tty::bold << setw(7) << setfill('0')
<< RoutingEvent::getProcesseds() << setfill(' ') << tty::reset << ">" << tty::cr;
cmess2.flush ();
} else {
cmess2 << " <repair.event:" << setw(7) << setfill('0')
<< RoutingEvent::getProcesseds() << setfill(' ') << " "
<< event->getEventLevel() << ":" << event->getPriority() << "> "
<< event->getSegment()
//<< "> @" << DbU::getValueString(event->getSegment()->getAxis())
//<< " id:" << event->getSegment()->getId()
//<< " " << event->getSegment()->getNet()->getName()
<< endl;
cmess2.flush();
}
event->process ( _eventQueue, _eventHistory, _eventLoop );
count++;
if ( RoutingEvent::getProcesseds() >= limit ) setInterrupt ( true );
}
// {
// ltrace(200) << (void*)event << " ["
// << (event->isCloned ()?"C":"-")
// << (event->isDisabled ()?"d":"-")
// << (event->isUnimplemented()?"u":"-") << "] "
// << event->getSegment() << endl;
// if ( not event->isCloned() and event->isUnimplemented() ) {
// count++;
// event->setProcessed ( false );
// event->setMode ( RoutingEvent::Repair );
// event->process ( _eventQueue, _eventHistory, _eventLoop );
// if (tty::enabled()) {
// cmess1 << " <event:"
// << tty::bold << tty::fgcolor(tty::Red) << setw(7) << setfill('0')
// << RoutingEvent::getProcesseds() << setfill(' ') << tty::reset << ">" << tty::cr;
// cmess1.flush ();
// } else {
// cmess1 << " <event:" << setw(7) << setfill('0')
// << RoutingEvent::getProcesseds() << setfill(' ') << ">" << endl;
// }
// }
// }
* ./hurricane/src/hviewer, ./coriolis/src/crlcore, ./coriolis/src/knik, ./coriolis/src/katabatic, ./coriolis/src/kite, ./coriolis/src/equinox, ./coriolis/src/solstice, ./coriolis/src/ispd: - SVN MOVE: Source tree simplification & uniformisation. Now all tools are at the same level, directly under the root of the repository. No more "coriolis/src".
2010-03-09 09:24:55 -06:00
if (count and tty::enabled()) cmess1 << endl;
size_t eventsCount = _eventHistory.size();
_eventHistory.clear();
_eventQueue.clear();
if ( RoutingEvent::getAllocateds() > 0 ) {
cerr << Bug("%d events remains after clear.",RoutingEvent::getAllocateds()) << endl;
}
// if ( _slowMotion && getCellWidget() ) {
// Session::close ();
// getCellWidget()->refresh();
// Session::open ( _kiteEngine );
// }
* ./kite: - Change: Big cleanup. Supress all the GCell level: GCell, GCellGrid, GCellRoutingSet. Blockages now managed inside the BuildPower. - New: In NegociateWindow, Histogram of GCell densities in Gnuplot format.
2010-12-12 15:42:57 -06:00
_statistics.setEventsCount ( eventsCount );
* ./hurricane/src/hviewer, ./coriolis/src/crlcore, ./coriolis/src/knik, ./coriolis/src/katabatic, ./coriolis/src/kite, ./coriolis/src/equinox, ./coriolis/src/solstice, ./coriolis/src/ispd: - SVN MOVE: Source tree simplification & uniformisation. Now all tools are at the same level, directly under the root of the repository. No more "coriolis/src".
2010-03-09 09:24:55 -06:00
ltraceout(149);
return eventsCount;
}
void NegociateWindow::run ( int slowMotion )
{
ltrace(150) << "NegociateWindow::run()" << endl;
ltracein(149);
* ./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
cmess1 << " o Running Negociate Algorithm" << endl;
* ./hurricane/src/hviewer, ./coriolis/src/crlcore, ./coriolis/src/knik, ./coriolis/src/katabatic, ./coriolis/src/kite, ./coriolis/src/equinox, ./coriolis/src/solstice, ./coriolis/src/ispd: - SVN MOVE: Source tree simplification & uniformisation. Now all tools are at the same level, directly under the root of the repository. No more "coriolis/src".
2010-03-09 09:24:55 -06:00
TrackElement::setOverlapCostCB ( NegociateOverlapCost );
RoutingEvent::resetProcesseds ();
* ./kite: - Change: Big cleanup. Supress all the GCell level: GCell, GCellGrid, GCellRoutingSet. Blockages now managed inside the BuildPower. - New: In NegociateWindow, Histogram of GCell densities in Gnuplot format.
2010-12-12 15:42:57 -06:00
for ( size_t igcell=0 ; igcell<_gcells.size() ; ++igcell ) {
_createRouting ( _gcells[igcell] );
}
Session::revalidate ();
getKiteEngine()->setMinimumWL ( computeWirelength() );
#if defined(CHECK_DATABASE)
unsigned int overlaps = 0;
Session::getKiteEngine()->_check(overlaps,"after _createRouting(GCell*)");
#endif
* ./hurricane/src/hviewer, ./coriolis/src/crlcore, ./coriolis/src/knik, ./coriolis/src/katabatic, ./coriolis/src/kite, ./coriolis/src/equinox, ./coriolis/src/solstice, ./coriolis/src/ispd: - SVN MOVE: Source tree simplification & uniformisation. Now all tools are at the same level, directly under the root of the repository. No more "coriolis/src".
2010-03-09 09:24:55 -06:00
_slowMotion = slowMotion;
* ./kite: - Change: Big cleanup. Supress all the GCell level: GCell, GCellGrid, GCellRoutingSet. Blockages now managed inside the BuildPower. - New: In NegociateWindow, Histogram of GCell densities in Gnuplot format.
2010-12-12 15:42:57 -06:00
_negociate ();
* ./hurricane/src/hviewer, ./coriolis/src/crlcore, ./coriolis/src/knik, ./coriolis/src/katabatic, ./coriolis/src/kite, ./coriolis/src/equinox, ./coriolis/src/solstice, ./coriolis/src/ispd: - SVN MOVE: Source tree simplification & uniformisation. Now all tools are at the same level, directly under the root of the repository. No more "coriolis/src".
2010-03-09 09:24:55 -06:00
Session::get()->isEmpty();
# if defined(CHECK_DATABASE)
_kite->_check ( overlaps, "after negociation" );
# endif
ltraceout(149);
}
void NegociateWindow::printStatistics () const
{
* ./katabatic, ./kite: - Bug: In Configuration, discrepencies of parameter types.
2010-06-22 08:59:22 -05:00
cmess1 << " o Computing statistics." << endl;
cmess1 << Dots::asSizet(" - Processeds Events Total",RoutingEvent::getProcesseds()) << endl;
cmess1 << Dots::asSizet(" - Unique Events Total"
,(RoutingEvent::getProcesseds() - RoutingEvent::getCloneds())) << endl;
* ./kite: - Change: Big cleanup. Supress all the GCell level: GCell, GCellGrid, GCellRoutingSet. Blockages now managed inside the BuildPower. - New: In NegociateWindow, Histogram of GCell densities in Gnuplot format.
2010-12-12 15:42:57 -06:00
cmess1 << Dots::asSizet(" - # of GCells",_statistics.getGCellsCount()) << endl;
* ./kite: - New: In Configuration & RoutingEvent, adds support for minimal break length by layer (METAL1 through METAL7). - New: Added measures for final detailed wirelength, wirelength expansion ratio, failed wirelength, number of processeds events and number of unique events.
2010-08-22 07:38:27 -05:00
addMeasure<size_t>( getCell(), "Events" , RoutingEvent::getProcesseds(), 12 );
addMeasure<size_t>( getCell(), "UEvents", RoutingEvent::getProcesseds()-RoutingEvent::getCloneds(), 12 );
* ./kite: - Change: Big cleanup. Supress all the GCell level: GCell, GCellGrid, GCellRoutingSet. Blockages now managed inside the BuildPower. - New: In NegociateWindow, Histogram of GCell densities in Gnuplot format.
2010-12-12 15:42:57 -06:00
* ./kite: - Change: Removed Histogram, moved into crlcore. - Change: In GraphicKiteEngine, add a menu entry to dump the measurments.
2010-12-15 09:12:00 -06:00
Histogram* densityHistogram = new Histogram ( 1.0, 0.1, 2 );
addMeasure<Histogram>( getCell(), "GCells Density Histogram", densityHistogram );
densityHistogram->setFileExtension ( ".density.histogram" );
densityHistogram->setMainTitle ( "GCell Densities" );
densityHistogram->setTitle ( "Avg. Density", 0 );
densityHistogram->setTitle ( "Peak Density", 1 );
densityHistogram->setColor ( "green", 0 );
densityHistogram->setColor ( "red" , 1 );
* ./kite: - Change: Big cleanup. Supress all the GCell level: GCell, GCellGrid, GCellRoutingSet. Blockages now managed inside the BuildPower. - New: In NegociateWindow, Histogram of GCell densities in Gnuplot format.
2010-12-12 15:42:57 -06:00
const Katabatic::GCellVector* gcells = getKiteEngine()->getGCellGrid()->getGCellVector();
* ./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
getKiteEngine()->getGCellGrid()->setDensityMode ( Katabatic::GCellGrid::MaxHVDensity );
* ./kite: - Change: Big cleanup. Supress all the GCell level: GCell, GCellGrid, GCellRoutingSet. Blockages now managed inside the BuildPower. - New: In NegociateWindow, Histogram of GCell densities in Gnuplot format.
2010-12-12 15:42:57 -06:00
for ( size_t igcell=0 ; igcell<(*gcells).size() ; ++igcell ) {
* ./kite: - Change: Removed Histogram, moved into crlcore. - Change: In GraphicKiteEngine, add a menu entry to dump the measurments.
2010-12-15 09:12:00 -06:00
densityHistogram->addSample ( (*gcells)[igcell]->getDensity(), 0 );
* ./kite: - Change: Big cleanup. Supress all the GCell level: GCell, GCellGrid, GCellRoutingSet. Blockages now managed inside the BuildPower. - New: In NegociateWindow, Histogram of GCell densities in Gnuplot format.
2010-12-12 15:42:57 -06:00
}
getKiteEngine()->getGCellGrid()->setDensityMode ( Katabatic::GCellGrid::MaxDensity );
for ( size_t igcell=0 ; igcell<(*gcells).size() ; ++igcell ) {
* ./kite: - Change: Removed Histogram, moved into crlcore. - Change: In GraphicKiteEngine, add a menu entry to dump the measurments.
2010-12-15 09:12:00 -06:00
densityHistogram->addSample ( (*gcells)[igcell]->getDensity(), 1 );
* ./kite: - Change: Big cleanup. Supress all the GCell level: GCell, GCellGrid, GCellRoutingSet. Blockages now managed inside the BuildPower. - New: In NegociateWindow, Histogram of GCell densities in Gnuplot format.
2010-12-12 15:42:57 -06:00
}
* ./kite: - Change: Removed Histogram, moved into crlcore. - Change: In GraphicKiteEngine, add a menu entry to dump the measurments.
2010-12-15 09:12:00 -06:00
densityHistogram->normalize ( 0 );
densityHistogram->normalize ( 1 );
* ./hurricane/src/hviewer, ./coriolis/src/crlcore, ./coriolis/src/knik, ./coriolis/src/katabatic, ./coriolis/src/kite, ./coriolis/src/equinox, ./coriolis/src/solstice, ./coriolis/src/ispd: - SVN MOVE: Source tree simplification & uniformisation. Now all tools are at the same level, directly under the root of the repository. No more "coriolis/src".
2010-03-09 09:24:55 -06:00
}
string NegociateWindow::_getString () const
{
ostringstream os;
* ./kite: - Change: Big cleanup. Supress all the GCell level: GCell, GCellGrid, GCellRoutingSet. Blockages now managed inside the BuildPower. - New: In NegociateWindow, Histogram of GCell densities in Gnuplot format.
2010-12-12 15:42:57 -06:00
os << "<" << _getTypeName() << ">";
* ./hurricane/src/hviewer, ./coriolis/src/crlcore, ./coriolis/src/knik, ./coriolis/src/katabatic, ./coriolis/src/kite, ./coriolis/src/equinox, ./coriolis/src/solstice, ./coriolis/src/ispd: - SVN MOVE: Source tree simplification & uniformisation. Now all tools are at the same level, directly under the root of the repository. No more "coriolis/src".
2010-03-09 09:24:55 -06:00
return ( os.str() );
}
Record* NegociateWindow::_getRecord () const
{
Record* record = new Record ( _getString() );
* ./kite: - Change: Big cleanup. Supress all the GCell level: GCell, GCellGrid, GCellRoutingSet. Blockages now managed inside the BuildPower. - New: In NegociateWindow, Histogram of GCell densities in Gnuplot format.
2010-12-12 15:42:57 -06:00
record->add ( getSlot ( "_gcells", _gcells ) );
* ./hurricane/src/hviewer, ./coriolis/src/crlcore, ./coriolis/src/knik, ./coriolis/src/katabatic, ./coriolis/src/kite, ./coriolis/src/equinox, ./coriolis/src/solstice, ./coriolis/src/ispd: - SVN MOVE: Source tree simplification & uniformisation. Now all tools are at the same level, directly under the root of the repository. No more "coriolis/src".
2010-03-09 09:24:55 -06:00
return ( record );
}
} // End of Kite namespace.