coriolis/etesian/src/EtesianEngine.cpp

1104 lines
38 KiB
C++
Raw Normal View History

Update to Qt 5, requires cmake 2.8.9. New placer: Etesian. Update to Qt 5: * Change: Now requires at least cmake 2.8.9. * Change: CMakeLists.txt needs small changes. Qt modules must be found one by one (Core, Gui, Widgets). Must add "set(CMAKE_AUTOMOC ON)" in the top file and replace "qt4" prefix in macros by "qt5". Added simpler macro "setup_qt()" in FindBootstrap.cmake. * Change: No longer need to include <QGtkStyle> is is choosen by default according to the current desktop environment. * Change: In <hurricane>, In HApplication, launch ExceptionWidget when a std::exception is catched instead of silently discarting it. New placer Tool: Etesian * New: <etesian> analytical placer. Encapsulate Coloquinte from Gabriel Gouvine. * New: in <documentation>, add stub demonstration ToolEngine <smurf>. Needs to be commented. Miscellaneous: * New: in <boostrap> and <unicorn>, added support for Etesian, the new analytic placer. The tool itself will be added in the next commit. * Bug: in <CellWidget>, when shifting the display buffer, we no longer can copy the buffer on itself (we should never have). Now go through a temporary one (PlaneId::AutoCopy) which is added to the DrawingPlanes. Affect "goLeft()" and "goUp()". * Bug: In <CellWidget>, remove the WA_PaintOnScreen flag/attribute. When it's on, no PaintEvent is transmitted to the CellWidget when it's the central widget of the <CellViewer> (QMainWindow). It's something I still don't understand from the doc of Qt. * Change: In <AreaCommand>, use the PlaneId enumeration instead of a anonymous numerical index. * Change: In <HApplication>, no longer catch and silently discard standartd exceptions but launch the ExceptionWidget... Suppress the now deprecated constructor with "Type" argument. * Change: In <SelectionModel>, the "reset()" method is deprecated in Qt5, instead enclose the "clear()" by a "beginResetModel()" and "endResetModel()" pair. * New: In <crlcore>, add support for ISPD05 benchmarks (in Bookshelf format). Forked from ISPD04 and not finished yet. * Change: In <Mauka>, distinguish the Action string identifier from <Etesian> * New: In <unicorn>, add entry for ISPD05 loader. Add entry for <Etesian> analytic placer.
2014-03-22 05:50:36 -05:00
// -*- C++ -*-
//
// This file is part of the Coriolis Software.
// Copyright (c) UPMC 2014-2018, All Rights Reserved
Update to Qt 5, requires cmake 2.8.9. New placer: Etesian. Update to Qt 5: * Change: Now requires at least cmake 2.8.9. * Change: CMakeLists.txt needs small changes. Qt modules must be found one by one (Core, Gui, Widgets). Must add "set(CMAKE_AUTOMOC ON)" in the top file and replace "qt4" prefix in macros by "qt5". Added simpler macro "setup_qt()" in FindBootstrap.cmake. * Change: No longer need to include <QGtkStyle> is is choosen by default according to the current desktop environment. * Change: In <hurricane>, In HApplication, launch ExceptionWidget when a std::exception is catched instead of silently discarting it. New placer Tool: Etesian * New: <etesian> analytical placer. Encapsulate Coloquinte from Gabriel Gouvine. * New: in <documentation>, add stub demonstration ToolEngine <smurf>. Needs to be commented. Miscellaneous: * New: in <boostrap> and <unicorn>, added support for Etesian, the new analytic placer. The tool itself will be added in the next commit. * Bug: in <CellWidget>, when shifting the display buffer, we no longer can copy the buffer on itself (we should never have). Now go through a temporary one (PlaneId::AutoCopy) which is added to the DrawingPlanes. Affect "goLeft()" and "goUp()". * Bug: In <CellWidget>, remove the WA_PaintOnScreen flag/attribute. When it's on, no PaintEvent is transmitted to the CellWidget when it's the central widget of the <CellViewer> (QMainWindow). It's something I still don't understand from the doc of Qt. * Change: In <AreaCommand>, use the PlaneId enumeration instead of a anonymous numerical index. * Change: In <HApplication>, no longer catch and silently discard standartd exceptions but launch the ExceptionWidget... Suppress the now deprecated constructor with "Type" argument. * Change: In <SelectionModel>, the "reset()" method is deprecated in Qt5, instead enclose the "clear()" by a "beginResetModel()" and "endResetModel()" pair. * New: In <crlcore>, add support for ISPD05 benchmarks (in Bookshelf format). Forked from ISPD04 and not finished yet. * Change: In <Mauka>, distinguish the Action string identifier from <Etesian> * New: In <unicorn>, add entry for ISPD05 loader. Add entry for <Etesian> analytic placer.
2014-03-22 05:50:36 -05:00
//
// +-----------------------------------------------------------------+
// | C O R I O L I S |
// | E t e s i a n - A n a l y t i c P l a c e r |
// | |
// | Author : Jean-Paul CHAPUT |
// | E-mail : Jean-Paul.Chaput@asim.lip6.fr |
// | =============================================================== |
// | C++ Module : "./EtesianEngine.cpp" |
// +-----------------------------------------------------------------+
Update to Qt 5, requires cmake 2.8.9. New placer: Etesian. Update to Qt 5: * Change: Now requires at least cmake 2.8.9. * Change: CMakeLists.txt needs small changes. Qt modules must be found one by one (Core, Gui, Widgets). Must add "set(CMAKE_AUTOMOC ON)" in the top file and replace "qt4" prefix in macros by "qt5". Added simpler macro "setup_qt()" in FindBootstrap.cmake. * Change: No longer need to include <QGtkStyle> is is choosen by default according to the current desktop environment. * Change: In <hurricane>, In HApplication, launch ExceptionWidget when a std::exception is catched instead of silently discarting it. New placer Tool: Etesian * New: <etesian> analytical placer. Encapsulate Coloquinte from Gabriel Gouvine. * New: in <documentation>, add stub demonstration ToolEngine <smurf>. Needs to be commented. Miscellaneous: * New: in <boostrap> and <unicorn>, added support for Etesian, the new analytic placer. The tool itself will be added in the next commit. * Bug: in <CellWidget>, when shifting the display buffer, we no longer can copy the buffer on itself (we should never have). Now go through a temporary one (PlaneId::AutoCopy) which is added to the DrawingPlanes. Affect "goLeft()" and "goUp()". * Bug: In <CellWidget>, remove the WA_PaintOnScreen flag/attribute. When it's on, no PaintEvent is transmitted to the CellWidget when it's the central widget of the <CellViewer> (QMainWindow). It's something I still don't understand from the doc of Qt. * Change: In <AreaCommand>, use the PlaneId enumeration instead of a anonymous numerical index. * Change: In <HApplication>, no longer catch and silently discard standartd exceptions but launch the ExceptionWidget... Suppress the now deprecated constructor with "Type" argument. * Change: In <SelectionModel>, the "reset()" method is deprecated in Qt5, instead enclose the "clear()" by a "beginResetModel()" and "endResetModel()" pair. * New: In <crlcore>, add support for ISPD05 benchmarks (in Bookshelf format). Forked from ISPD04 and not finished yet. * Change: In <Mauka>, distinguish the Action string identifier from <Etesian> * New: In <unicorn>, add entry for ISPD05 loader. Add entry for <Etesian> analytic placer.
2014-03-22 05:50:36 -05:00
#include <sstream>
#include <fstream>
#include <iomanip>
#include "coloquinte/circuit.hxx"
#include "coloquinte/legalizer.hxx"
#include "vlsisapd/configuration/Configuration.h"
#include "vlsisapd/utilities/Dots.h"
Update to Qt 5, requires cmake 2.8.9. New placer: Etesian. Update to Qt 5: * Change: Now requires at least cmake 2.8.9. * Change: CMakeLists.txt needs small changes. Qt modules must be found one by one (Core, Gui, Widgets). Must add "set(CMAKE_AUTOMOC ON)" in the top file and replace "qt4" prefix in macros by "qt5". Added simpler macro "setup_qt()" in FindBootstrap.cmake. * Change: No longer need to include <QGtkStyle> is is choosen by default according to the current desktop environment. * Change: In <hurricane>, In HApplication, launch ExceptionWidget when a std::exception is catched instead of silently discarting it. New placer Tool: Etesian * New: <etesian> analytical placer. Encapsulate Coloquinte from Gabriel Gouvine. * New: in <documentation>, add stub demonstration ToolEngine <smurf>. Needs to be commented. Miscellaneous: * New: in <boostrap> and <unicorn>, added support for Etesian, the new analytic placer. The tool itself will be added in the next commit. * Bug: in <CellWidget>, when shifting the display buffer, we no longer can copy the buffer on itself (we should never have). Now go through a temporary one (PlaneId::AutoCopy) which is added to the DrawingPlanes. Affect "goLeft()" and "goUp()". * Bug: In <CellWidget>, remove the WA_PaintOnScreen flag/attribute. When it's on, no PaintEvent is transmitted to the CellWidget when it's the central widget of the <CellViewer> (QMainWindow). It's something I still don't understand from the doc of Qt. * Change: In <AreaCommand>, use the PlaneId enumeration instead of a anonymous numerical index. * Change: In <HApplication>, no longer catch and silently discard standartd exceptions but launch the ExceptionWidget... Suppress the now deprecated constructor with "Type" argument. * Change: In <SelectionModel>, the "reset()" method is deprecated in Qt5, instead enclose the "clear()" by a "beginResetModel()" and "endResetModel()" pair. * New: In <crlcore>, add support for ISPD05 benchmarks (in Bookshelf format). Forked from ISPD04 and not finished yet. * Change: In <Mauka>, distinguish the Action string identifier from <Etesian> * New: In <unicorn>, add entry for ISPD05 loader. Add entry for <Etesian> analytic placer.
2014-03-22 05:50:36 -05:00
#include "hurricane/DebugSession.h"
#include "hurricane/Bug.h"
#include "hurricane/Error.h"
#include "hurricane/Warning.h"
#include "hurricane/Breakpoint.h"
#include "hurricane/DataBase.h"
Update to Qt 5, requires cmake 2.8.9. New placer: Etesian. Update to Qt 5: * Change: Now requires at least cmake 2.8.9. * Change: CMakeLists.txt needs small changes. Qt modules must be found one by one (Core, Gui, Widgets). Must add "set(CMAKE_AUTOMOC ON)" in the top file and replace "qt4" prefix in macros by "qt5". Added simpler macro "setup_qt()" in FindBootstrap.cmake. * Change: No longer need to include <QGtkStyle> is is choosen by default according to the current desktop environment. * Change: In <hurricane>, In HApplication, launch ExceptionWidget when a std::exception is catched instead of silently discarting it. New placer Tool: Etesian * New: <etesian> analytical placer. Encapsulate Coloquinte from Gabriel Gouvine. * New: in <documentation>, add stub demonstration ToolEngine <smurf>. Needs to be commented. Miscellaneous: * New: in <boostrap> and <unicorn>, added support for Etesian, the new analytic placer. The tool itself will be added in the next commit. * Bug: in <CellWidget>, when shifting the display buffer, we no longer can copy the buffer on itself (we should never have). Now go through a temporary one (PlaneId::AutoCopy) which is added to the DrawingPlanes. Affect "goLeft()" and "goUp()". * Bug: In <CellWidget>, remove the WA_PaintOnScreen flag/attribute. When it's on, no PaintEvent is transmitted to the CellWidget when it's the central widget of the <CellViewer> (QMainWindow). It's something I still don't understand from the doc of Qt. * Change: In <AreaCommand>, use the PlaneId enumeration instead of a anonymous numerical index. * Change: In <HApplication>, no longer catch and silently discard standartd exceptions but launch the ExceptionWidget... Suppress the now deprecated constructor with "Type" argument. * Change: In <SelectionModel>, the "reset()" method is deprecated in Qt5, instead enclose the "clear()" by a "beginResetModel()" and "endResetModel()" pair. * New: In <crlcore>, add support for ISPD05 benchmarks (in Bookshelf format). Forked from ISPD04 and not finished yet. * Change: In <Mauka>, distinguish the Action string identifier from <Etesian> * New: In <unicorn>, add entry for ISPD05 loader. Add entry for <Etesian> analytic placer.
2014-03-22 05:50:36 -05:00
#include "hurricane/Layer.h"
#include "hurricane/Net.h"
#include "hurricane/Pad.h"
#include "hurricane/Plug.h"
#include "hurricane/Cell.h"
#include "hurricane/Occurrence.h"
#include "hurricane/Instance.h"
#include "hurricane/Vertical.h"
#include "hurricane/Horizontal.h"
#include "hurricane/RoutingPad.h"
#include "hurricane/UpdateSession.h"
#include "hurricane/viewer/CellWidget.h"
Better handling of exceptions between C++ & Python. Misc. checks. * Bug: In Hurricane, in PyHurricane.h the macro HCATCH was not catching standard STL exceptions. This was the source of the cryptic message: "Fatal Python error: Py_EndInterpreter: thread still has a frame" The Python interpreter was interrupted uncleanly bypassing it's own exceptions mechanism. In PyViewer, the Viewer *do not* inherit from a base class (in the Python export). * New: In Hurricane, in DbU, compute maximum values (in double) for grid, lambda & physical (in meter) so now the DbU::toGrid(), DbU::toLambda() & DbU::toPhysical() methods can check for out of bound values, and throw an exception. * Change: In Hurricane, ExceptionWidget::catchAllWrapper() now returns a boolean, set to <true> if an exception has been catched. Allow callers to interrupt themselves if a problem has occured. * Bug: In Kite & Etesian, in the Python wrapper, send a Python exception if catchAllwrapper() did return true, instead of continuing... * Change: In Kite & Etesian, adds a setViewer() method (exported in Python) to use the graphical ExceptionWidget when in graphic mode. * Bug: In Cumulus, in PadsCorona.py the check for the core vs. chip size was not returning False when invalid. * New: In CRL Core, in Vst driver, add a support IEEE VHDL. Inactive for now as I don't see clearly the policy for selecting it or not. Remove the code of the old Vst driver. In Blif parser, check for non-existent models (incomplete or corrupted Blif file). Found by G. Gouvine. * New: Added extras file for IDE-like support under Emacs.
2015-05-20 07:02:18 -05:00
#include "hurricane/viewer/CellViewer.h"
#include "katabatic/GCellGrid.h"
#include "katabatic/KatabaticEngine.h"
2015-04-20 09:50:30 -05:00
#include "kite/KiteEngine.h"
Update to Qt 5, requires cmake 2.8.9. New placer: Etesian. Update to Qt 5: * Change: Now requires at least cmake 2.8.9. * Change: CMakeLists.txt needs small changes. Qt modules must be found one by one (Core, Gui, Widgets). Must add "set(CMAKE_AUTOMOC ON)" in the top file and replace "qt4" prefix in macros by "qt5". Added simpler macro "setup_qt()" in FindBootstrap.cmake. * Change: No longer need to include <QGtkStyle> is is choosen by default according to the current desktop environment. * Change: In <hurricane>, In HApplication, launch ExceptionWidget when a std::exception is catched instead of silently discarting it. New placer Tool: Etesian * New: <etesian> analytical placer. Encapsulate Coloquinte from Gabriel Gouvine. * New: in <documentation>, add stub demonstration ToolEngine <smurf>. Needs to be commented. Miscellaneous: * New: in <boostrap> and <unicorn>, added support for Etesian, the new analytic placer. The tool itself will be added in the next commit. * Bug: in <CellWidget>, when shifting the display buffer, we no longer can copy the buffer on itself (we should never have). Now go through a temporary one (PlaneId::AutoCopy) which is added to the DrawingPlanes. Affect "goLeft()" and "goUp()". * Bug: In <CellWidget>, remove the WA_PaintOnScreen flag/attribute. When it's on, no PaintEvent is transmitted to the CellWidget when it's the central widget of the <CellViewer> (QMainWindow). It's something I still don't understand from the doc of Qt. * Change: In <AreaCommand>, use the PlaneId enumeration instead of a anonymous numerical index. * Change: In <HApplication>, no longer catch and silently discard standartd exceptions but launch the ExceptionWidget... Suppress the now deprecated constructor with "Type" argument. * Change: In <SelectionModel>, the "reset()" method is deprecated in Qt5, instead enclose the "clear()" by a "beginResetModel()" and "endResetModel()" pair. * New: In <crlcore>, add support for ISPD05 benchmarks (in Bookshelf format). Forked from ISPD04 and not finished yet. * Change: In <Mauka>, distinguish the Action string identifier from <Etesian> * New: In <unicorn>, add entry for ISPD05 loader. Add entry for <Etesian> analytic placer.
2014-03-22 05:50:36 -05:00
#include "crlcore/Utilities.h"
#include "crlcore/Measures.h"
#include "crlcore/AllianceFramework.h"
#include "etesian/EtesianEngine.h"
#include "etesian/FeedCells.h"
Update to Qt 5, requires cmake 2.8.9. New placer: Etesian. Update to Qt 5: * Change: Now requires at least cmake 2.8.9. * Change: CMakeLists.txt needs small changes. Qt modules must be found one by one (Core, Gui, Widgets). Must add "set(CMAKE_AUTOMOC ON)" in the top file and replace "qt4" prefix in macros by "qt5". Added simpler macro "setup_qt()" in FindBootstrap.cmake. * Change: No longer need to include <QGtkStyle> is is choosen by default according to the current desktop environment. * Change: In <hurricane>, In HApplication, launch ExceptionWidget when a std::exception is catched instead of silently discarting it. New placer Tool: Etesian * New: <etesian> analytical placer. Encapsulate Coloquinte from Gabriel Gouvine. * New: in <documentation>, add stub demonstration ToolEngine <smurf>. Needs to be commented. Miscellaneous: * New: in <boostrap> and <unicorn>, added support for Etesian, the new analytic placer. The tool itself will be added in the next commit. * Bug: in <CellWidget>, when shifting the display buffer, we no longer can copy the buffer on itself (we should never have). Now go through a temporary one (PlaneId::AutoCopy) which is added to the DrawingPlanes. Affect "goLeft()" and "goUp()". * Bug: In <CellWidget>, remove the WA_PaintOnScreen flag/attribute. When it's on, no PaintEvent is transmitted to the CellWidget when it's the central widget of the <CellViewer> (QMainWindow). It's something I still don't understand from the doc of Qt. * Change: In <AreaCommand>, use the PlaneId enumeration instead of a anonymous numerical index. * Change: In <HApplication>, no longer catch and silently discard standartd exceptions but launch the ExceptionWidget... Suppress the now deprecated constructor with "Type" argument. * Change: In <SelectionModel>, the "reset()" method is deprecated in Qt5, instead enclose the "clear()" by a "beginResetModel()" and "endResetModel()" pair. * New: In <crlcore>, add support for ISPD05 benchmarks (in Bookshelf format). Forked from ISPD04 and not finished yet. * Change: In <Mauka>, distinguish the Action string identifier from <Etesian> * New: In <unicorn>, add entry for ISPD05 loader. Add entry for <Etesian> analytic placer.
2014-03-22 05:50:36 -05:00
namespace {
using namespace std;
using namespace Hurricane;
using coloquinte::int_t;
using coloquinte::float_t;
using coloquinte::point;
// Options for both placers
unsigned const SteinerModel = 0x0001;
// Options for the global placer
unsigned const ForceUniformDensity = 0x0010;
unsigned const UpdateLB = 0x0020;
unsigned const UpdateUB = 0x0040;
// Options for the detailed placer
unsigned const UpdateDetailed = 0x0100;
//unsigned const NonConvexOpt = 0x0200;
Update to Qt 5, requires cmake 2.8.9. New placer: Etesian. Update to Qt 5: * Change: Now requires at least cmake 2.8.9. * Change: CMakeLists.txt needs small changes. Qt modules must be found one by one (Core, Gui, Widgets). Must add "set(CMAKE_AUTOMOC ON)" in the top file and replace "qt4" prefix in macros by "qt5". Added simpler macro "setup_qt()" in FindBootstrap.cmake. * Change: No longer need to include <QGtkStyle> is is choosen by default according to the current desktop environment. * Change: In <hurricane>, In HApplication, launch ExceptionWidget when a std::exception is catched instead of silently discarting it. New placer Tool: Etesian * New: <etesian> analytical placer. Encapsulate Coloquinte from Gabriel Gouvine. * New: in <documentation>, add stub demonstration ToolEngine <smurf>. Needs to be commented. Miscellaneous: * New: in <boostrap> and <unicorn>, added support for Etesian, the new analytic placer. The tool itself will be added in the next commit. * Bug: in <CellWidget>, when shifting the display buffer, we no longer can copy the buffer on itself (we should never have). Now go through a temporary one (PlaneId::AutoCopy) which is added to the DrawingPlanes. Affect "goLeft()" and "goUp()". * Bug: In <CellWidget>, remove the WA_PaintOnScreen flag/attribute. When it's on, no PaintEvent is transmitted to the CellWidget when it's the central widget of the <CellViewer> (QMainWindow). It's something I still don't understand from the doc of Qt. * Change: In <AreaCommand>, use the PlaneId enumeration instead of a anonymous numerical index. * Change: In <HApplication>, no longer catch and silently discard standartd exceptions but launch the ExceptionWidget... Suppress the now deprecated constructor with "Type" argument. * Change: In <SelectionModel>, the "reset()" method is deprecated in Qt5, instead enclose the "clear()" by a "beginResetModel()" and "endResetModel()" pair. * New: In <crlcore>, add support for ISPD05 benchmarks (in Bookshelf format). Forked from ISPD04 and not finished yet. * Change: In <Mauka>, distinguish the Action string identifier from <Etesian> * New: In <unicorn>, add entry for ISPD05 loader. Add entry for <Etesian> analytic placer.
2014-03-22 05:50:36 -05:00
string extractInstanceName ( const RoutingPad* rp )
{
ostringstream name;
Occurrence occurrence = rp->getOccurrence();
name << getString(occurrence.getOwnerCell()->getName()) << ':';
if (not rp->getOccurrence().getPath().getHeadPath().isEmpty())
name << getString(rp->getOccurrence().getPath().getHeadPath().getName()) << ":";
if (rp->getOccurrence().getPath().getTailInstance())
name << "I." << getString(rp->getOccurrence().getPath().getTailInstance()->getName());
else
name << "C";
Update to Qt 5, requires cmake 2.8.9. New placer: Etesian. Update to Qt 5: * Change: Now requires at least cmake 2.8.9. * Change: CMakeLists.txt needs small changes. Qt modules must be found one by one (Core, Gui, Widgets). Must add "set(CMAKE_AUTOMOC ON)" in the top file and replace "qt4" prefix in macros by "qt5". Added simpler macro "setup_qt()" in FindBootstrap.cmake. * Change: No longer need to include <QGtkStyle> is is choosen by default according to the current desktop environment. * Change: In <hurricane>, In HApplication, launch ExceptionWidget when a std::exception is catched instead of silently discarting it. New placer Tool: Etesian * New: <etesian> analytical placer. Encapsulate Coloquinte from Gabriel Gouvine. * New: in <documentation>, add stub demonstration ToolEngine <smurf>. Needs to be commented. Miscellaneous: * New: in <boostrap> and <unicorn>, added support for Etesian, the new analytic placer. The tool itself will be added in the next commit. * Bug: in <CellWidget>, when shifting the display buffer, we no longer can copy the buffer on itself (we should never have). Now go through a temporary one (PlaneId::AutoCopy) which is added to the DrawingPlanes. Affect "goLeft()" and "goUp()". * Bug: In <CellWidget>, remove the WA_PaintOnScreen flag/attribute. When it's on, no PaintEvent is transmitted to the CellWidget when it's the central widget of the <CellViewer> (QMainWindow). It's something I still don't understand from the doc of Qt. * Change: In <AreaCommand>, use the PlaneId enumeration instead of a anonymous numerical index. * Change: In <HApplication>, no longer catch and silently discard standartd exceptions but launch the ExceptionWidget... Suppress the now deprecated constructor with "Type" argument. * Change: In <SelectionModel>, the "reset()" method is deprecated in Qt5, instead enclose the "clear()" by a "beginResetModel()" and "endResetModel()" pair. * New: In <crlcore>, add support for ISPD05 benchmarks (in Bookshelf format). Forked from ISPD04 and not finished yet. * Change: In <Mauka>, distinguish the Action string identifier from <Etesian> * New: In <unicorn>, add entry for ISPD05 loader. Add entry for <Etesian> analytic placer.
2014-03-22 05:50:36 -05:00
return name.str();
}
#if THIS_IS_DISABLED
Update to Qt 5, requires cmake 2.8.9. New placer: Etesian. Update to Qt 5: * Change: Now requires at least cmake 2.8.9. * Change: CMakeLists.txt needs small changes. Qt modules must be found one by one (Core, Gui, Widgets). Must add "set(CMAKE_AUTOMOC ON)" in the top file and replace "qt4" prefix in macros by "qt5". Added simpler macro "setup_qt()" in FindBootstrap.cmake. * Change: No longer need to include <QGtkStyle> is is choosen by default according to the current desktop environment. * Change: In <hurricane>, In HApplication, launch ExceptionWidget when a std::exception is catched instead of silently discarting it. New placer Tool: Etesian * New: <etesian> analytical placer. Encapsulate Coloquinte from Gabriel Gouvine. * New: in <documentation>, add stub demonstration ToolEngine <smurf>. Needs to be commented. Miscellaneous: * New: in <boostrap> and <unicorn>, added support for Etesian, the new analytic placer. The tool itself will be added in the next commit. * Bug: in <CellWidget>, when shifting the display buffer, we no longer can copy the buffer on itself (we should never have). Now go through a temporary one (PlaneId::AutoCopy) which is added to the DrawingPlanes. Affect "goLeft()" and "goUp()". * Bug: In <CellWidget>, remove the WA_PaintOnScreen flag/attribute. When it's on, no PaintEvent is transmitted to the CellWidget when it's the central widget of the <CellViewer> (QMainWindow). It's something I still don't understand from the doc of Qt. * Change: In <AreaCommand>, use the PlaneId enumeration instead of a anonymous numerical index. * Change: In <HApplication>, no longer catch and silently discard standartd exceptions but launch the ExceptionWidget... Suppress the now deprecated constructor with "Type" argument. * Change: In <SelectionModel>, the "reset()" method is deprecated in Qt5, instead enclose the "clear()" by a "beginResetModel()" and "endResetModel()" pair. * New: In <crlcore>, add support for ISPD05 benchmarks (in Bookshelf format). Forked from ISPD04 and not finished yet. * Change: In <Mauka>, distinguish the Action string identifier from <Etesian> * New: In <unicorn>, add entry for ISPD05 loader. Add entry for <Etesian> analytic placer.
2014-03-22 05:50:36 -05:00
string extractPinName ( const RoutingPad* rp )
{
ostringstream name;
Occurrence occurrence = rp->getOccurrence();
name << getString(occurrence.getOwnerCell()->getName()) << ':';
if (not rp->getOccurrence().getPath().isEmpty())
name << getString(rp->getOccurrence().getPath().getName()) << ":";
name << "T." << getString(rp->_getEntityAsComponent()->getNet()->getName());
return name.str();
}
string extractTerminalName ( const RoutingPad* rp )
{
ostringstream name;
Occurrence occurrence = rp->getOccurrence();
name << getString(occurrence.getOwnerCell()->getName()) << ':';
if (not rp->getOccurrence().getPath().isEmpty())
name << getString(rp->getOccurrence().getPath().getName()) << ":";
name << "T." << getString(rp->_getEntityAsComponent()->getNet()->getName());
return name.str();
}
#endif
Update to Qt 5, requires cmake 2.8.9. New placer: Etesian. Update to Qt 5: * Change: Now requires at least cmake 2.8.9. * Change: CMakeLists.txt needs small changes. Qt modules must be found one by one (Core, Gui, Widgets). Must add "set(CMAKE_AUTOMOC ON)" in the top file and replace "qt4" prefix in macros by "qt5". Added simpler macro "setup_qt()" in FindBootstrap.cmake. * Change: No longer need to include <QGtkStyle> is is choosen by default according to the current desktop environment. * Change: In <hurricane>, In HApplication, launch ExceptionWidget when a std::exception is catched instead of silently discarting it. New placer Tool: Etesian * New: <etesian> analytical placer. Encapsulate Coloquinte from Gabriel Gouvine. * New: in <documentation>, add stub demonstration ToolEngine <smurf>. Needs to be commented. Miscellaneous: * New: in <boostrap> and <unicorn>, added support for Etesian, the new analytic placer. The tool itself will be added in the next commit. * Bug: in <CellWidget>, when shifting the display buffer, we no longer can copy the buffer on itself (we should never have). Now go through a temporary one (PlaneId::AutoCopy) which is added to the DrawingPlanes. Affect "goLeft()" and "goUp()". * Bug: In <CellWidget>, remove the WA_PaintOnScreen flag/attribute. When it's on, no PaintEvent is transmitted to the CellWidget when it's the central widget of the <CellViewer> (QMainWindow). It's something I still don't understand from the doc of Qt. * Change: In <AreaCommand>, use the PlaneId enumeration instead of a anonymous numerical index. * Change: In <HApplication>, no longer catch and silently discard standartd exceptions but launch the ExceptionWidget... Suppress the now deprecated constructor with "Type" argument. * Change: In <SelectionModel>, the "reset()" method is deprecated in Qt5, instead enclose the "clear()" by a "beginResetModel()" and "endResetModel()" pair. * New: In <crlcore>, add support for ISPD05 benchmarks (in Bookshelf format). Forked from ISPD04 and not finished yet. * Change: In <Mauka>, distinguish the Action string identifier from <Etesian> * New: In <unicorn>, add entry for ISPD05 loader. Add entry for <Etesian> analytic placer.
2014-03-22 05:50:36 -05:00
Point extractRpOffset ( const RoutingPad* rp )
{
Cell* masterCell = rp->getOccurrence().getMasterCell();
Component* component = rp->_getEntityAsComponent();
// TODO: verify that it doesn't assume that the orientation is North
Box masterBox = masterCell->getAbutmentBox();
Update to Qt 5, requires cmake 2.8.9. New placer: Etesian. Update to Qt 5: * Change: Now requires at least cmake 2.8.9. * Change: CMakeLists.txt needs small changes. Qt modules must be found one by one (Core, Gui, Widgets). Must add "set(CMAKE_AUTOMOC ON)" in the top file and replace "qt4" prefix in macros by "qt5". Added simpler macro "setup_qt()" in FindBootstrap.cmake. * Change: No longer need to include <QGtkStyle> is is choosen by default according to the current desktop environment. * Change: In <hurricane>, In HApplication, launch ExceptionWidget when a std::exception is catched instead of silently discarting it. New placer Tool: Etesian * New: <etesian> analytical placer. Encapsulate Coloquinte from Gabriel Gouvine. * New: in <documentation>, add stub demonstration ToolEngine <smurf>. Needs to be commented. Miscellaneous: * New: in <boostrap> and <unicorn>, added support for Etesian, the new analytic placer. The tool itself will be added in the next commit. * Bug: in <CellWidget>, when shifting the display buffer, we no longer can copy the buffer on itself (we should never have). Now go through a temporary one (PlaneId::AutoCopy) which is added to the DrawingPlanes. Affect "goLeft()" and "goUp()". * Bug: In <CellWidget>, remove the WA_PaintOnScreen flag/attribute. When it's on, no PaintEvent is transmitted to the CellWidget when it's the central widget of the <CellViewer> (QMainWindow). It's something I still don't understand from the doc of Qt. * Change: In <AreaCommand>, use the PlaneId enumeration instead of a anonymous numerical index. * Change: In <HApplication>, no longer catch and silently discard standartd exceptions but launch the ExceptionWidget... Suppress the now deprecated constructor with "Type" argument. * Change: In <SelectionModel>, the "reset()" method is deprecated in Qt5, instead enclose the "clear()" by a "beginResetModel()" and "endResetModel()" pair. * New: In <crlcore>, add support for ISPD05 benchmarks (in Bookshelf format). Forked from ISPD04 and not finished yet. * Change: In <Mauka>, distinguish the Action string identifier from <Etesian> * New: In <unicorn>, add entry for ISPD05 loader. Add entry for <Etesian> analytic placer.
2014-03-22 05:50:36 -05:00
Point offset;
Update to Qt 5, requires cmake 2.8.9. New placer: Etesian. Update to Qt 5: * Change: Now requires at least cmake 2.8.9. * Change: CMakeLists.txt needs small changes. Qt modules must be found one by one (Core, Gui, Widgets). Must add "set(CMAKE_AUTOMOC ON)" in the top file and replace "qt4" prefix in macros by "qt5". Added simpler macro "setup_qt()" in FindBootstrap.cmake. * Change: No longer need to include <QGtkStyle> is is choosen by default according to the current desktop environment. * Change: In <hurricane>, In HApplication, launch ExceptionWidget when a std::exception is catched instead of silently discarting it. New placer Tool: Etesian * New: <etesian> analytical placer. Encapsulate Coloquinte from Gabriel Gouvine. * New: in <documentation>, add stub demonstration ToolEngine <smurf>. Needs to be commented. Miscellaneous: * New: in <boostrap> and <unicorn>, added support for Etesian, the new analytic placer. The tool itself will be added in the next commit. * Bug: in <CellWidget>, when shifting the display buffer, we no longer can copy the buffer on itself (we should never have). Now go through a temporary one (PlaneId::AutoCopy) which is added to the DrawingPlanes. Affect "goLeft()" and "goUp()". * Bug: In <CellWidget>, remove the WA_PaintOnScreen flag/attribute. When it's on, no PaintEvent is transmitted to the CellWidget when it's the central widget of the <CellViewer> (QMainWindow). It's something I still don't understand from the doc of Qt. * Change: In <AreaCommand>, use the PlaneId enumeration instead of a anonymous numerical index. * Change: In <HApplication>, no longer catch and silently discard standartd exceptions but launch the ExceptionWidget... Suppress the now deprecated constructor with "Type" argument. * Change: In <SelectionModel>, the "reset()" method is deprecated in Qt5, instead enclose the "clear()" by a "beginResetModel()" and "endResetModel()" pair. * New: In <crlcore>, add support for ISPD05 benchmarks (in Bookshelf format). Forked from ISPD04 and not finished yet. * Change: In <Mauka>, distinguish the Action string identifier from <Etesian> * New: In <unicorn>, add entry for ISPD05 loader. Add entry for <Etesian> analytic placer.
2014-03-22 05:50:36 -05:00
if (component) {
offset.setX( component->getCenter().getX() - masterBox.getXMin() );
offset.setY( component->getCenter().getY() - masterBox.getYMin() );
}
else { // Why?
offset = masterBox.getCenter();
Update to Qt 5, requires cmake 2.8.9. New placer: Etesian. Update to Qt 5: * Change: Now requires at least cmake 2.8.9. * Change: CMakeLists.txt needs small changes. Qt modules must be found one by one (Core, Gui, Widgets). Must add "set(CMAKE_AUTOMOC ON)" in the top file and replace "qt4" prefix in macros by "qt5". Added simpler macro "setup_qt()" in FindBootstrap.cmake. * Change: No longer need to include <QGtkStyle> is is choosen by default according to the current desktop environment. * Change: In <hurricane>, In HApplication, launch ExceptionWidget when a std::exception is catched instead of silently discarting it. New placer Tool: Etesian * New: <etesian> analytical placer. Encapsulate Coloquinte from Gabriel Gouvine. * New: in <documentation>, add stub demonstration ToolEngine <smurf>. Needs to be commented. Miscellaneous: * New: in <boostrap> and <unicorn>, added support for Etesian, the new analytic placer. The tool itself will be added in the next commit. * Bug: in <CellWidget>, when shifting the display buffer, we no longer can copy the buffer on itself (we should never have). Now go through a temporary one (PlaneId::AutoCopy) which is added to the DrawingPlanes. Affect "goLeft()" and "goUp()". * Bug: In <CellWidget>, remove the WA_PaintOnScreen flag/attribute. When it's on, no PaintEvent is transmitted to the CellWidget when it's the central widget of the <CellViewer> (QMainWindow). It's something I still don't understand from the doc of Qt. * Change: In <AreaCommand>, use the PlaneId enumeration instead of a anonymous numerical index. * Change: In <HApplication>, no longer catch and silently discard standartd exceptions but launch the ExceptionWidget... Suppress the now deprecated constructor with "Type" argument. * Change: In <SelectionModel>, the "reset()" method is deprecated in Qt5, instead enclose the "clear()" by a "beginResetModel()" and "endResetModel()" pair. * New: In <crlcore>, add support for ISPD05 benchmarks (in Bookshelf format). Forked from ISPD04 and not finished yet. * Change: In <Mauka>, distinguish the Action string identifier from <Etesian> * New: In <unicorn>, add entry for ISPD05 loader. Add entry for <Etesian> analytic placer.
2014-03-22 05:50:36 -05:00
}
return offset;
}
Transformation toTransformation ( point<int_t> position
, point<bool> orientation
, Cell* model
, DbU::Unit hpitch
, DbU::Unit vpitch
)
{
DbU::Unit tx = position.x * vpitch;
DbU::Unit ty = position.y * hpitch;
Box cellBox = model->getAbutmentBox();
Transformation::Orientation orient = Transformation::Orientation::ID;
if ( not orientation.x and orientation.y) {
tx += cellBox.getWidth();
orient = Transformation::Orientation::MX;
} else if ( orientation.x and not orientation.y) {
ty += cellBox.getHeight();
orient = Transformation::Orientation::MY;
} else if ( not orientation.x and not orientation.y) {
tx += cellBox.getWidth();
ty += cellBox.getHeight();
orient = Transformation::Orientation::R2;
}
return Transformation( tx, ty, orient );
}
Update to Qt 5, requires cmake 2.8.9. New placer: Etesian. Update to Qt 5: * Change: Now requires at least cmake 2.8.9. * Change: CMakeLists.txt needs small changes. Qt modules must be found one by one (Core, Gui, Widgets). Must add "set(CMAKE_AUTOMOC ON)" in the top file and replace "qt4" prefix in macros by "qt5". Added simpler macro "setup_qt()" in FindBootstrap.cmake. * Change: No longer need to include <QGtkStyle> is is choosen by default according to the current desktop environment. * Change: In <hurricane>, In HApplication, launch ExceptionWidget when a std::exception is catched instead of silently discarting it. New placer Tool: Etesian * New: <etesian> analytical placer. Encapsulate Coloquinte from Gabriel Gouvine. * New: in <documentation>, add stub demonstration ToolEngine <smurf>. Needs to be commented. Miscellaneous: * New: in <boostrap> and <unicorn>, added support for Etesian, the new analytic placer. The tool itself will be added in the next commit. * Bug: in <CellWidget>, when shifting the display buffer, we no longer can copy the buffer on itself (we should never have). Now go through a temporary one (PlaneId::AutoCopy) which is added to the DrawingPlanes. Affect "goLeft()" and "goUp()". * Bug: In <CellWidget>, remove the WA_PaintOnScreen flag/attribute. When it's on, no PaintEvent is transmitted to the CellWidget when it's the central widget of the <CellViewer> (QMainWindow). It's something I still don't understand from the doc of Qt. * Change: In <AreaCommand>, use the PlaneId enumeration instead of a anonymous numerical index. * Change: In <HApplication>, no longer catch and silently discard standartd exceptions but launch the ExceptionWidget... Suppress the now deprecated constructor with "Type" argument. * Change: In <SelectionModel>, the "reset()" method is deprecated in Qt5, instead enclose the "clear()" by a "beginResetModel()" and "endResetModel()" pair. * New: In <crlcore>, add support for ISPD05 benchmarks (in Bookshelf format). Forked from ISPD04 and not finished yet. * Change: In <Mauka>, distinguish the Action string identifier from <Etesian> * New: In <unicorn>, add entry for ISPD05 loader. Add entry for <Etesian> analytic placer.
2014-03-22 05:50:36 -05:00
} // Anonymous namespace.
namespace Etesian {
using std::cout;
using std::cerr;
using std::endl;
using std::setw;
using std::left;
using std::string;
using std::ostream;
using std::ofstream;
using std::ostringstream;
using std::setprecision;
using std::vector;
using std::pair;
using std::make_pair;
using std::unordered_map;
using Utilities::Dots;
Update to Qt 5, requires cmake 2.8.9. New placer: Etesian. Update to Qt 5: * Change: Now requires at least cmake 2.8.9. * Change: CMakeLists.txt needs small changes. Qt modules must be found one by one (Core, Gui, Widgets). Must add "set(CMAKE_AUTOMOC ON)" in the top file and replace "qt4" prefix in macros by "qt5". Added simpler macro "setup_qt()" in FindBootstrap.cmake. * Change: No longer need to include <QGtkStyle> is is choosen by default according to the current desktop environment. * Change: In <hurricane>, In HApplication, launch ExceptionWidget when a std::exception is catched instead of silently discarting it. New placer Tool: Etesian * New: <etesian> analytical placer. Encapsulate Coloquinte from Gabriel Gouvine. * New: in <documentation>, add stub demonstration ToolEngine <smurf>. Needs to be commented. Miscellaneous: * New: in <boostrap> and <unicorn>, added support for Etesian, the new analytic placer. The tool itself will be added in the next commit. * Bug: in <CellWidget>, when shifting the display buffer, we no longer can copy the buffer on itself (we should never have). Now go through a temporary one (PlaneId::AutoCopy) which is added to the DrawingPlanes. Affect "goLeft()" and "goUp()". * Bug: In <CellWidget>, remove the WA_PaintOnScreen flag/attribute. When it's on, no PaintEvent is transmitted to the CellWidget when it's the central widget of the <CellViewer> (QMainWindow). It's something I still don't understand from the doc of Qt. * Change: In <AreaCommand>, use the PlaneId enumeration instead of a anonymous numerical index. * Change: In <HApplication>, no longer catch and silently discard standartd exceptions but launch the ExceptionWidget... Suppress the now deprecated constructor with "Type" argument. * Change: In <SelectionModel>, the "reset()" method is deprecated in Qt5, instead enclose the "clear()" by a "beginResetModel()" and "endResetModel()" pair. * New: In <crlcore>, add support for ISPD05 benchmarks (in Bookshelf format). Forked from ISPD04 and not finished yet. * Change: In <Mauka>, distinguish the Action string identifier from <Etesian> * New: In <unicorn>, add entry for ISPD05 loader. Add entry for <Etesian> analytic placer.
2014-03-22 05:50:36 -05:00
using Hurricane::DebugSession;
using Hurricane::tab;
using Hurricane::ForEachIterator;
using Hurricane::Bug;
using Hurricane::Error;
using Hurricane::Warning;
using Hurricane::Breakpoint;
using Hurricane::Box;
using Hurricane::DataBase;
Update to Qt 5, requires cmake 2.8.9. New placer: Etesian. Update to Qt 5: * Change: Now requires at least cmake 2.8.9. * Change: CMakeLists.txt needs small changes. Qt modules must be found one by one (Core, Gui, Widgets). Must add "set(CMAKE_AUTOMOC ON)" in the top file and replace "qt4" prefix in macros by "qt5". Added simpler macro "setup_qt()" in FindBootstrap.cmake. * Change: No longer need to include <QGtkStyle> is is choosen by default according to the current desktop environment. * Change: In <hurricane>, In HApplication, launch ExceptionWidget when a std::exception is catched instead of silently discarting it. New placer Tool: Etesian * New: <etesian> analytical placer. Encapsulate Coloquinte from Gabriel Gouvine. * New: in <documentation>, add stub demonstration ToolEngine <smurf>. Needs to be commented. Miscellaneous: * New: in <boostrap> and <unicorn>, added support for Etesian, the new analytic placer. The tool itself will be added in the next commit. * Bug: in <CellWidget>, when shifting the display buffer, we no longer can copy the buffer on itself (we should never have). Now go through a temporary one (PlaneId::AutoCopy) which is added to the DrawingPlanes. Affect "goLeft()" and "goUp()". * Bug: In <CellWidget>, remove the WA_PaintOnScreen flag/attribute. When it's on, no PaintEvent is transmitted to the CellWidget when it's the central widget of the <CellViewer> (QMainWindow). It's something I still don't understand from the doc of Qt. * Change: In <AreaCommand>, use the PlaneId enumeration instead of a anonymous numerical index. * Change: In <HApplication>, no longer catch and silently discard standartd exceptions but launch the ExceptionWidget... Suppress the now deprecated constructor with "Type" argument. * Change: In <SelectionModel>, the "reset()" method is deprecated in Qt5, instead enclose the "clear()" by a "beginResetModel()" and "endResetModel()" pair. * New: In <crlcore>, add support for ISPD05 benchmarks (in Bookshelf format). Forked from ISPD04 and not finished yet. * Change: In <Mauka>, distinguish the Action string identifier from <Etesian> * New: In <unicorn>, add entry for ISPD05 loader. Add entry for <Etesian> analytic placer.
2014-03-22 05:50:36 -05:00
using Hurricane::Layer;
using Hurricane::Cell;
using Hurricane::Instance;
using Hurricane::RoutingPad;
using Hurricane::Net;
using Hurricane::Occurrence;
using Hurricane::CellWidget;
Update to Qt 5, requires cmake 2.8.9. New placer: Etesian. Update to Qt 5: * Change: Now requires at least cmake 2.8.9. * Change: CMakeLists.txt needs small changes. Qt modules must be found one by one (Core, Gui, Widgets). Must add "set(CMAKE_AUTOMOC ON)" in the top file and replace "qt4" prefix in macros by "qt5". Added simpler macro "setup_qt()" in FindBootstrap.cmake. * Change: No longer need to include <QGtkStyle> is is choosen by default according to the current desktop environment. * Change: In <hurricane>, In HApplication, launch ExceptionWidget when a std::exception is catched instead of silently discarting it. New placer Tool: Etesian * New: <etesian> analytical placer. Encapsulate Coloquinte from Gabriel Gouvine. * New: in <documentation>, add stub demonstration ToolEngine <smurf>. Needs to be commented. Miscellaneous: * New: in <boostrap> and <unicorn>, added support for Etesian, the new analytic placer. The tool itself will be added in the next commit. * Bug: in <CellWidget>, when shifting the display buffer, we no longer can copy the buffer on itself (we should never have). Now go through a temporary one (PlaneId::AutoCopy) which is added to the DrawingPlanes. Affect "goLeft()" and "goUp()". * Bug: In <CellWidget>, remove the WA_PaintOnScreen flag/attribute. When it's on, no PaintEvent is transmitted to the CellWidget when it's the central widget of the <CellViewer> (QMainWindow). It's something I still don't understand from the doc of Qt. * Change: In <AreaCommand>, use the PlaneId enumeration instead of a anonymous numerical index. * Change: In <HApplication>, no longer catch and silently discard standartd exceptions but launch the ExceptionWidget... Suppress the now deprecated constructor with "Type" argument. * Change: In <SelectionModel>, the "reset()" method is deprecated in Qt5, instead enclose the "clear()" by a "beginResetModel()" and "endResetModel()" pair. * New: In <crlcore>, add support for ISPD05 benchmarks (in Bookshelf format). Forked from ISPD04 and not finished yet. * Change: In <Mauka>, distinguish the Action string identifier from <Etesian> * New: In <unicorn>, add entry for ISPD05 loader. Add entry for <Etesian> analytic placer.
2014-03-22 05:50:36 -05:00
using CRL::ToolEngine;
using CRL::AllianceFramework;
using CRL::Catalog;
Update to Qt 5, requires cmake 2.8.9. New placer: Etesian. Update to Qt 5: * Change: Now requires at least cmake 2.8.9. * Change: CMakeLists.txt needs small changes. Qt modules must be found one by one (Core, Gui, Widgets). Must add "set(CMAKE_AUTOMOC ON)" in the top file and replace "qt4" prefix in macros by "qt5". Added simpler macro "setup_qt()" in FindBootstrap.cmake. * Change: No longer need to include <QGtkStyle> is is choosen by default according to the current desktop environment. * Change: In <hurricane>, In HApplication, launch ExceptionWidget when a std::exception is catched instead of silently discarting it. New placer Tool: Etesian * New: <etesian> analytical placer. Encapsulate Coloquinte from Gabriel Gouvine. * New: in <documentation>, add stub demonstration ToolEngine <smurf>. Needs to be commented. Miscellaneous: * New: in <boostrap> and <unicorn>, added support for Etesian, the new analytic placer. The tool itself will be added in the next commit. * Bug: in <CellWidget>, when shifting the display buffer, we no longer can copy the buffer on itself (we should never have). Now go through a temporary one (PlaneId::AutoCopy) which is added to the DrawingPlanes. Affect "goLeft()" and "goUp()". * Bug: In <CellWidget>, remove the WA_PaintOnScreen flag/attribute. When it's on, no PaintEvent is transmitted to the CellWidget when it's the central widget of the <CellViewer> (QMainWindow). It's something I still don't understand from the doc of Qt. * Change: In <AreaCommand>, use the PlaneId enumeration instead of a anonymous numerical index. * Change: In <HApplication>, no longer catch and silently discard standartd exceptions but launch the ExceptionWidget... Suppress the now deprecated constructor with "Type" argument. * Change: In <SelectionModel>, the "reset()" method is deprecated in Qt5, instead enclose the "clear()" by a "beginResetModel()" and "endResetModel()" pair. * New: In <crlcore>, add support for ISPD05 benchmarks (in Bookshelf format). Forked from ISPD04 and not finished yet. * Change: In <Mauka>, distinguish the Action string identifier from <Etesian> * New: In <unicorn>, add entry for ISPD05 loader. Add entry for <Etesian> analytic placer.
2014-03-22 05:50:36 -05:00
using CRL::addMeasure;
using CRL::Measures;
using CRL::MeasuresSet;
using CRL::CatalogExtension;
using coloquinte::index_t;
using coloquinte::capacity_t;
using coloquinte::int_t;
using coloquinte::float_t;
using coloquinte::point;
using coloquinte::box;
using coloquinte::Movability;
using coloquinte::temporary_cell;
using coloquinte::temporary_net;
using coloquinte::temporary_pin;
using coloquinte::netlist;
using coloquinte::placement_t;
Update to Qt 5, requires cmake 2.8.9. New placer: Etesian. Update to Qt 5: * Change: Now requires at least cmake 2.8.9. * Change: CMakeLists.txt needs small changes. Qt modules must be found one by one (Core, Gui, Widgets). Must add "set(CMAKE_AUTOMOC ON)" in the top file and replace "qt4" prefix in macros by "qt5". Added simpler macro "setup_qt()" in FindBootstrap.cmake. * Change: No longer need to include <QGtkStyle> is is choosen by default according to the current desktop environment. * Change: In <hurricane>, In HApplication, launch ExceptionWidget when a std::exception is catched instead of silently discarting it. New placer Tool: Etesian * New: <etesian> analytical placer. Encapsulate Coloquinte from Gabriel Gouvine. * New: in <documentation>, add stub demonstration ToolEngine <smurf>. Needs to be commented. Miscellaneous: * New: in <boostrap> and <unicorn>, added support for Etesian, the new analytic placer. The tool itself will be added in the next commit. * Bug: in <CellWidget>, when shifting the display buffer, we no longer can copy the buffer on itself (we should never have). Now go through a temporary one (PlaneId::AutoCopy) which is added to the DrawingPlanes. Affect "goLeft()" and "goUp()". * Bug: In <CellWidget>, remove the WA_PaintOnScreen flag/attribute. When it's on, no PaintEvent is transmitted to the CellWidget when it's the central widget of the <CellViewer> (QMainWindow). It's something I still don't understand from the doc of Qt. * Change: In <AreaCommand>, use the PlaneId enumeration instead of a anonymous numerical index. * Change: In <HApplication>, no longer catch and silently discard standartd exceptions but launch the ExceptionWidget... Suppress the now deprecated constructor with "Type" argument. * Change: In <SelectionModel>, the "reset()" method is deprecated in Qt5, instead enclose the "clear()" by a "beginResetModel()" and "endResetModel()" pair. * New: In <crlcore>, add support for ISPD05 benchmarks (in Bookshelf format). Forked from ISPD04 and not finished yet. * Change: In <Mauka>, distinguish the Action string identifier from <Etesian> * New: In <unicorn>, add entry for ISPD05 loader. Add entry for <Etesian> analytic placer.
2014-03-22 05:50:36 -05:00
const char* missingEtesian =
"%s :\n\n"
" Cell %s do not have any EtesianEngine (or not yet created).\n";
// -------------------------------------------------------------------
// Class : "Etesian::EtesianEngine".
Name EtesianEngine::_toolName = "Etesian";
const Name& EtesianEngine::staticGetName ()
{ return _toolName; }
EtesianEngine* EtesianEngine::get ( const Cell* cell )
{ return static_cast<EtesianEngine*>(ToolEngine::get(cell,staticGetName())); }
EtesianEngine::EtesianEngine ( Cell* cell )
Apaired segments building for symmetric routing (step 1). * Change: In Hurricane::BaseFlags, store flags in uint64_t instead of unsigned int because we start to need more than 32 different flags in some tools. * New: In ::getString() & ::getRecord() templates, add support for std::array<>. * Change: In CRL::ToolEngine, add support for timer (time & memory measurements) displaced from Katabatic. This way all ToolEngine can use this feature. The _postCreate() method display the memory just after ToolEngine allocation. * Change: In Etesian::EtesianEngine, make use of the ToolEngine builtin timer (remove the local one). Forgot to call the base class _postCreate() and _preDestroy(). * Change: In Anabatic::AnabaticEngine, make use of the ToolEngine builtin timer (remove the local one). * New: In Anabatic, new AutoSegments_Connecteds() collection. This Collection allows a deterministic walkthough *all* the AutoSegments connected either to source or target of one AutoSegment. * New: In Anabatic::AutoContactTerminal::isEndPoint() to check if an AutoContactTerminal is the *only one* anchored on a RoutingPad, thus being a true "end point" and not a kind of feed-through. * New: In Katana::KatanaEngine, added support for symmetric nets. Created new class DataSymmetric to store symmetric information of a net (mainly the paired AutoSegments). Added KatanaEngine::runSymmetricRouter(), for now only build the DataSymmetric informations. More to come... * Change: In Katana::GraphicKatanaEngine::_runTest(), now perform symmetric information building the non-symmetric routing.
2017-03-12 13:34:12 -05:00
: Super (cell)
, _configuration(new Configuration())
, _placed (false)
, _flatDesign (false)
, _surface ()
, _circuit ()
, _placementLB ()
, _placementUB ()
, _cellsToIds ()
, _idsToInsts ()
Better handling of exceptions between C++ & Python. Misc. checks. * Bug: In Hurricane, in PyHurricane.h the macro HCATCH was not catching standard STL exceptions. This was the source of the cryptic message: "Fatal Python error: Py_EndInterpreter: thread still has a frame" The Python interpreter was interrupted uncleanly bypassing it's own exceptions mechanism. In PyViewer, the Viewer *do not* inherit from a base class (in the Python export). * New: In Hurricane, in DbU, compute maximum values (in double) for grid, lambda & physical (in meter) so now the DbU::toGrid(), DbU::toLambda() & DbU::toPhysical() methods can check for out of bound values, and throw an exception. * Change: In Hurricane, ExceptionWidget::catchAllWrapper() now returns a boolean, set to <true> if an exception has been catched. Allow callers to interrupt themselves if a problem has occured. * Bug: In Kite & Etesian, in the Python wrapper, send a Python exception if catchAllwrapper() did return true, instead of continuing... * Change: In Kite & Etesian, adds a setViewer() method (exported in Python) to use the graphical ExceptionWidget when in graphic mode. * Bug: In Cumulus, in PadsCorona.py the check for the core vs. chip size was not returning False when invalid. * New: In CRL Core, in Vst driver, add a support IEEE VHDL. Inactive for now as I don't see clearly the policy for selecting it or not. Remove the code of the old Vst driver. In Blif parser, check for non-existent models (incomplete or corrupted Blif file). Found by G. Gouvine. * New: Added extras file for IDE-like support under Emacs.
2015-05-20 07:02:18 -05:00
, _viewer (NULL)
, _feedCells (this)
{
}
Update to Qt 5, requires cmake 2.8.9. New placer: Etesian. Update to Qt 5: * Change: Now requires at least cmake 2.8.9. * Change: CMakeLists.txt needs small changes. Qt modules must be found one by one (Core, Gui, Widgets). Must add "set(CMAKE_AUTOMOC ON)" in the top file and replace "qt4" prefix in macros by "qt5". Added simpler macro "setup_qt()" in FindBootstrap.cmake. * Change: No longer need to include <QGtkStyle> is is choosen by default according to the current desktop environment. * Change: In <hurricane>, In HApplication, launch ExceptionWidget when a std::exception is catched instead of silently discarting it. New placer Tool: Etesian * New: <etesian> analytical placer. Encapsulate Coloquinte from Gabriel Gouvine. * New: in <documentation>, add stub demonstration ToolEngine <smurf>. Needs to be commented. Miscellaneous: * New: in <boostrap> and <unicorn>, added support for Etesian, the new analytic placer. The tool itself will be added in the next commit. * Bug: in <CellWidget>, when shifting the display buffer, we no longer can copy the buffer on itself (we should never have). Now go through a temporary one (PlaneId::AutoCopy) which is added to the DrawingPlanes. Affect "goLeft()" and "goUp()". * Bug: In <CellWidget>, remove the WA_PaintOnScreen flag/attribute. When it's on, no PaintEvent is transmitted to the CellWidget when it's the central widget of the <CellViewer> (QMainWindow). It's something I still don't understand from the doc of Qt. * Change: In <AreaCommand>, use the PlaneId enumeration instead of a anonymous numerical index. * Change: In <HApplication>, no longer catch and silently discard standartd exceptions but launch the ExceptionWidget... Suppress the now deprecated constructor with "Type" argument. * Change: In <SelectionModel>, the "reset()" method is deprecated in Qt5, instead enclose the "clear()" by a "beginResetModel()" and "endResetModel()" pair. * New: In <crlcore>, add support for ISPD05 benchmarks (in Bookshelf format). Forked from ISPD04 and not finished yet. * Change: In <Mauka>, distinguish the Action string identifier from <Etesian> * New: In <unicorn>, add entry for ISPD05 loader. Add entry for <Etesian> analytic placer.
2014-03-22 05:50:36 -05:00
void EtesianEngine::_postCreate ()
{
Apaired segments building for symmetric routing (step 1). * Change: In Hurricane::BaseFlags, store flags in uint64_t instead of unsigned int because we start to need more than 32 different flags in some tools. * New: In ::getString() & ::getRecord() templates, add support for std::array<>. * Change: In CRL::ToolEngine, add support for timer (time & memory measurements) displaced from Katabatic. This way all ToolEngine can use this feature. The _postCreate() method display the memory just after ToolEngine allocation. * Change: In Etesian::EtesianEngine, make use of the ToolEngine builtin timer (remove the local one). Forgot to call the base class _postCreate() and _preDestroy(). * Change: In Anabatic::AnabaticEngine, make use of the ToolEngine builtin timer (remove the local one). * New: In Anabatic, new AutoSegments_Connecteds() collection. This Collection allows a deterministic walkthough *all* the AutoSegments connected either to source or target of one AutoSegment. * New: In Anabatic::AutoContactTerminal::isEndPoint() to check if an AutoContactTerminal is the *only one* anchored on a RoutingPad, thus being a true "end point" and not a kind of feed-through. * New: In Katana::KatanaEngine, added support for symmetric nets. Created new class DataSymmetric to store symmetric information of a net (mainly the paired AutoSegments). Added KatanaEngine::runSymmetricRouter(), for now only build the DataSymmetric informations. More to come... * Change: In Katana::GraphicKatanaEngine::_runTest(), now perform symmetric information building the non-symmetric routing.
2017-03-12 13:34:12 -05:00
Super::_postCreate();
Etesian integration with Chip & ClockTree plugins. * New: In Hurricane, in DebugSession, add a new method to activate the trace inconditionally with a certain level. * New: In Hurricane, in HyperNet, allow copy construction as there is no reason to disallow it and we need it now. * New: In Hurricane, in Cell::flattenNets(), add a new option to prevent the flattening of the clock net(s). For more safety perform the DeepNet creation *outside* the Collection loop. * Bug: In Hurricane, in Cell_HyperNetRootNetOccurrences, skip the DeepNets because they are the result of another flattening operation. * New: In Isobar, in PyBasicLayer, export C++ method getBlockageLayer(). * New: In Isobar, in PyRoutingGauge, export C++ method getLayerPitch(), needed by Cumulus plugins. * New: In Etesian, EtesianEngine::findYSpin() to look for the Y orientation if some cells are already placed. Typically the buffers of a clock tree. Pass the correct orienation to row_compatible_orientation(). Do not try to add feeds in the ISPD05 benchmarks. For now the benchmarks are detected through their names (unreliable). * Change: In Knik, in KnikEngine::initGlobalRouting(), allow the clock to be routed as an ordinary signal when the clock tree is not used. * New: In Kite, in BuildPowerRails, management & detection for the pre-routed clock. In KiteEngine constructor, early initialization of the blockage net to avoid later troubles in BuildPowerRails. * New: In Cumulus, in ChipPlugin, add support for Etesian plus new configuration parameter 'clockTree.placerEngine' to select between Mauka and Etesian. * New: In Cumulus, in BlockCorona, add blockages in the vertical sides in the vertical layer to prevent the router to use the vertical tracks in under the prower lines (for example, blockage in M3 with power line M5). In Cumulus, in ChipConf add attribute to access the blockage net. * New: In Cumulus, when the clock tree is disabled, do not generate the last rail around the block (the clock rail). * Bug: In Cumulus, in ChipConf use the clock pad to guess the clock signals and *not* the power pad. Add more meaningful error messages if a pad global signal is not found (implicit connexion by name). * Bug: In Cumulus, in ClockTree, compute correctly the cells Y spin, that is *from the bottom of the cell AB* (not from 'zero').
2015-02-24 06:13:17 -06:00
// Ugly: Name based detection of ISPD benchmarks.
if (getString(getCell()->getName()).substr(0,7) == "bigblue") {
cmess2 << " o ISPD benchmark <" << getCell()->getName()
Etesian integration with Chip & ClockTree plugins. * New: In Hurricane, in DebugSession, add a new method to activate the trace inconditionally with a certain level. * New: In Hurricane, in HyperNet, allow copy construction as there is no reason to disallow it and we need it now. * New: In Hurricane, in Cell::flattenNets(), add a new option to prevent the flattening of the clock net(s). For more safety perform the DeepNet creation *outside* the Collection loop. * Bug: In Hurricane, in Cell_HyperNetRootNetOccurrences, skip the DeepNets because they are the result of another flattening operation. * New: In Isobar, in PyBasicLayer, export C++ method getBlockageLayer(). * New: In Isobar, in PyRoutingGauge, export C++ method getLayerPitch(), needed by Cumulus plugins. * New: In Etesian, EtesianEngine::findYSpin() to look for the Y orientation if some cells are already placed. Typically the buffers of a clock tree. Pass the correct orienation to row_compatible_orientation(). Do not try to add feeds in the ISPD05 benchmarks. For now the benchmarks are detected through their names (unreliable). * Change: In Knik, in KnikEngine::initGlobalRouting(), allow the clock to be routed as an ordinary signal when the clock tree is not used. * New: In Kite, in BuildPowerRails, management & detection for the pre-routed clock. In KiteEngine constructor, early initialization of the blockage net to avoid later troubles in BuildPowerRails. * New: In Cumulus, in ChipPlugin, add support for Etesian plus new configuration parameter 'clockTree.placerEngine' to select between Mauka and Etesian. * New: In Cumulus, in BlockCorona, add blockages in the vertical sides in the vertical layer to prevent the router to use the vertical tracks in under the prower lines (for example, blockage in M3 with power line M5). In Cumulus, in ChipConf add attribute to access the blockage net. * New: In Cumulus, when the clock tree is disabled, do not generate the last rail around the block (the clock rail). * Bug: In Cumulus, in ChipConf use the clock pad to guess the clock signals and *not* the power pad. Add more meaningful error messages if a pad global signal is not found (implicit connexion by name). * Bug: In Cumulus, in ClockTree, compute correctly the cells Y spin, that is *from the bottom of the cell AB* (not from 'zero').
2015-02-24 06:13:17 -06:00
<< ">, no feed cells will be added." << endl;
} else {
string feedNames = getConfiguration()->getFeedNames();
char separator = ',';
while ( not feedNames.empty() ) {
size_t cut = feedNames.find( separator );
string feedName;
if (cut != string::npos) {
feedName = feedNames.substr( 0, cut );
feedNames = feedNames.substr( cut+1 );
} else {
feedName = feedNames;
feedNames.clear();
}
Cell* feed = DataBase::getDB()->getCell( feedName );
if (not feed)
feed = AllianceFramework::get()->getCell( feedName, Catalog::State::Views|Catalog::State::Foreign );
if (feed)
_feedCells.useFeed( feed );
else
cerr << Warning( "EtesianEngine::_postCreate() Unable to find \"%s\" feed cell."
, feedName.c_str()
) << endl;
}
Etesian integration with Chip & ClockTree plugins. * New: In Hurricane, in DebugSession, add a new method to activate the trace inconditionally with a certain level. * New: In Hurricane, in HyperNet, allow copy construction as there is no reason to disallow it and we need it now. * New: In Hurricane, in Cell::flattenNets(), add a new option to prevent the flattening of the clock net(s). For more safety perform the DeepNet creation *outside* the Collection loop. * Bug: In Hurricane, in Cell_HyperNetRootNetOccurrences, skip the DeepNets because they are the result of another flattening operation. * New: In Isobar, in PyBasicLayer, export C++ method getBlockageLayer(). * New: In Isobar, in PyRoutingGauge, export C++ method getLayerPitch(), needed by Cumulus plugins. * New: In Etesian, EtesianEngine::findYSpin() to look for the Y orientation if some cells are already placed. Typically the buffers of a clock tree. Pass the correct orienation to row_compatible_orientation(). Do not try to add feeds in the ISPD05 benchmarks. For now the benchmarks are detected through their names (unreliable). * Change: In Knik, in KnikEngine::initGlobalRouting(), allow the clock to be routed as an ordinary signal when the clock tree is not used. * New: In Kite, in BuildPowerRails, management & detection for the pre-routed clock. In KiteEngine constructor, early initialization of the blockage net to avoid later troubles in BuildPowerRails. * New: In Cumulus, in ChipPlugin, add support for Etesian plus new configuration parameter 'clockTree.placerEngine' to select between Mauka and Etesian. * New: In Cumulus, in BlockCorona, add blockages in the vertical sides in the vertical layer to prevent the router to use the vertical tracks in under the prower lines (for example, blockage in M3 with power line M5). In Cumulus, in ChipConf add attribute to access the blockage net. * New: In Cumulus, when the clock tree is disabled, do not generate the last rail around the block (the clock rail). * Bug: In Cumulus, in ChipConf use the clock pad to guess the clock signals and *not* the power pad. Add more meaningful error messages if a pad global signal is not found (implicit connexion by name). * Bug: In Cumulus, in ClockTree, compute correctly the cells Y spin, that is *from the bottom of the cell AB* (not from 'zero').
2015-02-24 06:13:17 -06:00
}
}
Update to Qt 5, requires cmake 2.8.9. New placer: Etesian. Update to Qt 5: * Change: Now requires at least cmake 2.8.9. * Change: CMakeLists.txt needs small changes. Qt modules must be found one by one (Core, Gui, Widgets). Must add "set(CMAKE_AUTOMOC ON)" in the top file and replace "qt4" prefix in macros by "qt5". Added simpler macro "setup_qt()" in FindBootstrap.cmake. * Change: No longer need to include <QGtkStyle> is is choosen by default according to the current desktop environment. * Change: In <hurricane>, In HApplication, launch ExceptionWidget when a std::exception is catched instead of silently discarting it. New placer Tool: Etesian * New: <etesian> analytical placer. Encapsulate Coloquinte from Gabriel Gouvine. * New: in <documentation>, add stub demonstration ToolEngine <smurf>. Needs to be commented. Miscellaneous: * New: in <boostrap> and <unicorn>, added support for Etesian, the new analytic placer. The tool itself will be added in the next commit. * Bug: in <CellWidget>, when shifting the display buffer, we no longer can copy the buffer on itself (we should never have). Now go through a temporary one (PlaneId::AutoCopy) which is added to the DrawingPlanes. Affect "goLeft()" and "goUp()". * Bug: In <CellWidget>, remove the WA_PaintOnScreen flag/attribute. When it's on, no PaintEvent is transmitted to the CellWidget when it's the central widget of the <CellViewer> (QMainWindow). It's something I still don't understand from the doc of Qt. * Change: In <AreaCommand>, use the PlaneId enumeration instead of a anonymous numerical index. * Change: In <HApplication>, no longer catch and silently discard standartd exceptions but launch the ExceptionWidget... Suppress the now deprecated constructor with "Type" argument. * Change: In <SelectionModel>, the "reset()" method is deprecated in Qt5, instead enclose the "clear()" by a "beginResetModel()" and "endResetModel()" pair. * New: In <crlcore>, add support for ISPD05 benchmarks (in Bookshelf format). Forked from ISPD04 and not finished yet. * Change: In <Mauka>, distinguish the Action string identifier from <Etesian> * New: In <unicorn>, add entry for ISPD05 loader. Add entry for <Etesian> analytic placer.
2014-03-22 05:50:36 -05:00
EtesianEngine* EtesianEngine::create ( Cell* cell )
{
EtesianEngine* etesian = new EtesianEngine ( cell );
etesian->_postCreate();
return etesian;
}
void EtesianEngine::_preDestroy ()
{
cdebug_log(129,1) << "EtesianEngine::_preDestroy()" << endl;
Update to Qt 5, requires cmake 2.8.9. New placer: Etesian. Update to Qt 5: * Change: Now requires at least cmake 2.8.9. * Change: CMakeLists.txt needs small changes. Qt modules must be found one by one (Core, Gui, Widgets). Must add "set(CMAKE_AUTOMOC ON)" in the top file and replace "qt4" prefix in macros by "qt5". Added simpler macro "setup_qt()" in FindBootstrap.cmake. * Change: No longer need to include <QGtkStyle> is is choosen by default according to the current desktop environment. * Change: In <hurricane>, In HApplication, launch ExceptionWidget when a std::exception is catched instead of silently discarting it. New placer Tool: Etesian * New: <etesian> analytical placer. Encapsulate Coloquinte from Gabriel Gouvine. * New: in <documentation>, add stub demonstration ToolEngine <smurf>. Needs to be commented. Miscellaneous: * New: in <boostrap> and <unicorn>, added support for Etesian, the new analytic placer. The tool itself will be added in the next commit. * Bug: in <CellWidget>, when shifting the display buffer, we no longer can copy the buffer on itself (we should never have). Now go through a temporary one (PlaneId::AutoCopy) which is added to the DrawingPlanes. Affect "goLeft()" and "goUp()". * Bug: In <CellWidget>, remove the WA_PaintOnScreen flag/attribute. When it's on, no PaintEvent is transmitted to the CellWidget when it's the central widget of the <CellViewer> (QMainWindow). It's something I still don't understand from the doc of Qt. * Change: In <AreaCommand>, use the PlaneId enumeration instead of a anonymous numerical index. * Change: In <HApplication>, no longer catch and silently discard standartd exceptions but launch the ExceptionWidget... Suppress the now deprecated constructor with "Type" argument. * Change: In <SelectionModel>, the "reset()" method is deprecated in Qt5, instead enclose the "clear()" by a "beginResetModel()" and "endResetModel()" pair. * New: In <crlcore>, add support for ISPD05 benchmarks (in Bookshelf format). Forked from ISPD04 and not finished yet. * Change: In <Mauka>, distinguish the Action string identifier from <Etesian> * New: In <unicorn>, add entry for ISPD05 loader. Add entry for <Etesian> analytic placer.
2014-03-22 05:50:36 -05:00
cmess1 << " o Deleting ToolEngine<" << getName() << "> from Cell <"
<< getCell()->getName() << ">" << endl;
Apaired segments building for symmetric routing (step 1). * Change: In Hurricane::BaseFlags, store flags in uint64_t instead of unsigned int because we start to need more than 32 different flags in some tools. * New: In ::getString() & ::getRecord() templates, add support for std::array<>. * Change: In CRL::ToolEngine, add support for timer (time & memory measurements) displaced from Katabatic. This way all ToolEngine can use this feature. The _postCreate() method display the memory just after ToolEngine allocation. * Change: In Etesian::EtesianEngine, make use of the ToolEngine builtin timer (remove the local one). Forgot to call the base class _postCreate() and _preDestroy(). * Change: In Anabatic::AnabaticEngine, make use of the ToolEngine builtin timer (remove the local one). * New: In Anabatic, new AutoSegments_Connecteds() collection. This Collection allows a deterministic walkthough *all* the AutoSegments connected either to source or target of one AutoSegment. * New: In Anabatic::AutoContactTerminal::isEndPoint() to check if an AutoContactTerminal is the *only one* anchored on a RoutingPad, thus being a true "end point" and not a kind of feed-through. * New: In Katana::KatanaEngine, added support for symmetric nets. Created new class DataSymmetric to store symmetric information of a net (mainly the paired AutoSegments). Added KatanaEngine::runSymmetricRouter(), for now only build the DataSymmetric informations. More to come... * Change: In Katana::GraphicKatanaEngine::_runTest(), now perform symmetric information building the non-symmetric routing.
2017-03-12 13:34:12 -05:00
Super::_preDestroy();
cdebug.log(129,-1);
Update to Qt 5, requires cmake 2.8.9. New placer: Etesian. Update to Qt 5: * Change: Now requires at least cmake 2.8.9. * Change: CMakeLists.txt needs small changes. Qt modules must be found one by one (Core, Gui, Widgets). Must add "set(CMAKE_AUTOMOC ON)" in the top file and replace "qt4" prefix in macros by "qt5". Added simpler macro "setup_qt()" in FindBootstrap.cmake. * Change: No longer need to include <QGtkStyle> is is choosen by default according to the current desktop environment. * Change: In <hurricane>, In HApplication, launch ExceptionWidget when a std::exception is catched instead of silently discarting it. New placer Tool: Etesian * New: <etesian> analytical placer. Encapsulate Coloquinte from Gabriel Gouvine. * New: in <documentation>, add stub demonstration ToolEngine <smurf>. Needs to be commented. Miscellaneous: * New: in <boostrap> and <unicorn>, added support for Etesian, the new analytic placer. The tool itself will be added in the next commit. * Bug: in <CellWidget>, when shifting the display buffer, we no longer can copy the buffer on itself (we should never have). Now go through a temporary one (PlaneId::AutoCopy) which is added to the DrawingPlanes. Affect "goLeft()" and "goUp()". * Bug: In <CellWidget>, remove the WA_PaintOnScreen flag/attribute. When it's on, no PaintEvent is transmitted to the CellWidget when it's the central widget of the <CellViewer> (QMainWindow). It's something I still don't understand from the doc of Qt. * Change: In <AreaCommand>, use the PlaneId enumeration instead of a anonymous numerical index. * Change: In <HApplication>, no longer catch and silently discard standartd exceptions but launch the ExceptionWidget... Suppress the now deprecated constructor with "Type" argument. * Change: In <SelectionModel>, the "reset()" method is deprecated in Qt5, instead enclose the "clear()" by a "beginResetModel()" and "endResetModel()" pair. * New: In <crlcore>, add support for ISPD05 benchmarks (in Bookshelf format). Forked from ISPD04 and not finished yet. * Change: In <Mauka>, distinguish the Action string identifier from <Etesian> * New: In <unicorn>, add entry for ISPD05 loader. Add entry for <Etesian> analytic placer.
2014-03-22 05:50:36 -05:00
}
EtesianEngine::~EtesianEngine ()
{
delete _configuration;
}
const Name& EtesianEngine::getName () const
{ return _toolName; }
const Configuration* EtesianEngine::getConfiguration () const
{ return _configuration; }
Update to Qt 5, requires cmake 2.8.9. New placer: Etesian. Update to Qt 5: * Change: Now requires at least cmake 2.8.9. * Change: CMakeLists.txt needs small changes. Qt modules must be found one by one (Core, Gui, Widgets). Must add "set(CMAKE_AUTOMOC ON)" in the top file and replace "qt4" prefix in macros by "qt5". Added simpler macro "setup_qt()" in FindBootstrap.cmake. * Change: No longer need to include <QGtkStyle> is is choosen by default according to the current desktop environment. * Change: In <hurricane>, In HApplication, launch ExceptionWidget when a std::exception is catched instead of silently discarting it. New placer Tool: Etesian * New: <etesian> analytical placer. Encapsulate Coloquinte from Gabriel Gouvine. * New: in <documentation>, add stub demonstration ToolEngine <smurf>. Needs to be commented. Miscellaneous: * New: in <boostrap> and <unicorn>, added support for Etesian, the new analytic placer. The tool itself will be added in the next commit. * Bug: in <CellWidget>, when shifting the display buffer, we no longer can copy the buffer on itself (we should never have). Now go through a temporary one (PlaneId::AutoCopy) which is added to the DrawingPlanes. Affect "goLeft()" and "goUp()". * Bug: In <CellWidget>, remove the WA_PaintOnScreen flag/attribute. When it's on, no PaintEvent is transmitted to the CellWidget when it's the central widget of the <CellViewer> (QMainWindow). It's something I still don't understand from the doc of Qt. * Change: In <AreaCommand>, use the PlaneId enumeration instead of a anonymous numerical index. * Change: In <HApplication>, no longer catch and silently discard standartd exceptions but launch the ExceptionWidget... Suppress the now deprecated constructor with "Type" argument. * Change: In <SelectionModel>, the "reset()" method is deprecated in Qt5, instead enclose the "clear()" by a "beginResetModel()" and "endResetModel()" pair. * New: In <crlcore>, add support for ISPD05 benchmarks (in Bookshelf format). Forked from ISPD04 and not finished yet. * Change: In <Mauka>, distinguish the Action string identifier from <Etesian> * New: In <unicorn>, add entry for ISPD05 loader. Add entry for <Etesian> analytic placer.
2014-03-22 05:50:36 -05:00
Configuration* EtesianEngine::getConfiguration ()
{ return _configuration; }
void EtesianEngine::setDefaultAb ()
{
double spaceMargin = getSpaceMargin();
double aspectRatio = getAspectRatio();
size_t instanceNb = 0;
double cellLength = 0;
vector<Occurrence> feedOccurrences;
for( Occurrence occurrence : getCell()->getLeafInstanceOccurrences() )
{
Instance* instance = static_cast<Instance*>(occurrence.getEntity());
Cell* masterCell = instance->getMasterCell();
string instanceName = occurrence.getCompactString();
if (CatalogExtension::isFeed(masterCell)) {
cerr << Warning( "Found a feedcell %s in an unplaced design, removing."
, instanceName.c_str()
) << endl;
feedOccurrences.push_back( occurrence );
continue;
}
cellLength += DbU::toLambda( masterCell->getAbutmentBox().getWidth() );
instanceNb += 1;
}
double gcellLength = cellLength*(1.0+spaceMargin) / DbU::toLambda( getSliceHeight() );
double rows = std::ceil( sqrt( gcellLength/aspectRatio ) );
double columns = std::ceil( gcellLength / rows );
UpdateSession::open();
for ( auto ioccurrence : feedOccurrences ) {
static_cast<Instance*>(ioccurrence.getEntity())->destroy();
}
DbU::Unit abWidth = columns*getSliceHeight();
DbU::Unit adjust = abWidth % getVerticalPitch();
if (adjust) abWidth += getVerticalPitch() - adjust;
getCell()->setAbutmentBox( Box( DbU::fromLambda(0)
, DbU::fromLambda(0)
, abWidth
, rows*getSliceHeight()
) );
UpdateSession::close();
Better handling of exceptions between C++ & Python. Misc. checks. * Bug: In Hurricane, in PyHurricane.h the macro HCATCH was not catching standard STL exceptions. This was the source of the cryptic message: "Fatal Python error: Py_EndInterpreter: thread still has a frame" The Python interpreter was interrupted uncleanly bypassing it's own exceptions mechanism. In PyViewer, the Viewer *do not* inherit from a base class (in the Python export). * New: In Hurricane, in DbU, compute maximum values (in double) for grid, lambda & physical (in meter) so now the DbU::toGrid(), DbU::toLambda() & DbU::toPhysical() methods can check for out of bound values, and throw an exception. * Change: In Hurricane, ExceptionWidget::catchAllWrapper() now returns a boolean, set to <true> if an exception has been catched. Allow callers to interrupt themselves if a problem has occured. * Bug: In Kite & Etesian, in the Python wrapper, send a Python exception if catchAllwrapper() did return true, instead of continuing... * Change: In Kite & Etesian, adds a setViewer() method (exported in Python) to use the graphical ExceptionWidget when in graphic mode. * Bug: In Cumulus, in PadsCorona.py the check for the core vs. chip size was not returning False when invalid. * New: In CRL Core, in Vst driver, add a support IEEE VHDL. Inactive for now as I don't see clearly the policy for selecting it or not. Remove the code of the old Vst driver. In Blif parser, check for non-existent models (incomplete or corrupted Blif file). Found by G. Gouvine. * New: Added extras file for IDE-like support under Emacs.
2015-05-20 07:02:18 -05:00
if (_viewer) _viewer->getCellWidget()->fitToContents();
cmess1 << " o Creating abutment box (margin:" << (spaceMargin*100.0)
<< "% aspect ratio:" << (aspectRatio*100.0)
<< "% g-length:" << (cellLength/DbU::toLambda(getSliceHeight()))
<< ")" << endl;
cmess1 << " - " << getCell()->getAbutmentBox() << endl;
cmess1 << " - GCell grid: [" << (int)columns << "x" << (int)rows << "]" << endl;
}
void EtesianEngine::resetPlacement ()
{
//cerr << "EtesianEngine::resetPlacement()" << endl;
if (not _placed) return;
_flatDesign = true;
Dots dots ( cmess2, " ", 80, 1000 );
if (not cmess2.enabled()) dots.disable();
cmess1 << " o Erasing previous placement of <" << getCell()->getName() << ">" << endl;
UpdateSession::open();
vector<Occurrence> feedOccurrences;
for( Occurrence occurrence : getCell()->getLeafInstanceOccurrences() )
{
dots.dot();
if ( _flatDesign and not occurrence.getPath().getTailPath().isEmpty())
_flatDesign = true;
Instance* instance = static_cast<Instance*>(occurrence.getEntity());
Cell* masterCell = instance->getMasterCell();
string instanceName = occurrence.getCompactString();
if (CatalogExtension::isFeed(masterCell)) {
feedOccurrences.push_back( occurrence );
}
}
for ( auto ioccurrence : feedOccurrences ) {
cerr << " Destroy: " << ioccurrence.getCompactString() << endl;
Instance* instance = static_cast<Instance*>(ioccurrence.getEntity());
instance->destroy();
}
UpdateSession::close();
dots.finish( Dots::Reset );
Better handling of exceptions between C++ & Python. Misc. checks. * Bug: In Hurricane, in PyHurricane.h the macro HCATCH was not catching standard STL exceptions. This was the source of the cryptic message: "Fatal Python error: Py_EndInterpreter: thread still has a frame" The Python interpreter was interrupted uncleanly bypassing it's own exceptions mechanism. In PyViewer, the Viewer *do not* inherit from a base class (in the Python export). * New: In Hurricane, in DbU, compute maximum values (in double) for grid, lambda & physical (in meter) so now the DbU::toGrid(), DbU::toLambda() & DbU::toPhysical() methods can check for out of bound values, and throw an exception. * Change: In Hurricane, ExceptionWidget::catchAllWrapper() now returns a boolean, set to <true> if an exception has been catched. Allow callers to interrupt themselves if a problem has occured. * Bug: In Kite & Etesian, in the Python wrapper, send a Python exception if catchAllwrapper() did return true, instead of continuing... * Change: In Kite & Etesian, adds a setViewer() method (exported in Python) to use the graphical ExceptionWidget when in graphic mode. * Bug: In Cumulus, in PadsCorona.py the check for the core vs. chip size was not returning False when invalid. * New: In CRL Core, in Vst driver, add a support IEEE VHDL. Inactive for now as I don't see clearly the policy for selecting it or not. Remove the code of the old Vst driver. In Blif parser, check for non-existent models (incomplete or corrupted Blif file). Found by G. Gouvine. * New: Added extras file for IDE-like support under Emacs.
2015-05-20 07:02:18 -05:00
if (_viewer) _viewer->getCellWidget()->refresh();
_placed = false;
}
void EtesianEngine::toColoquinte ()
Update to Qt 5, requires cmake 2.8.9. New placer: Etesian. Update to Qt 5: * Change: Now requires at least cmake 2.8.9. * Change: CMakeLists.txt needs small changes. Qt modules must be found one by one (Core, Gui, Widgets). Must add "set(CMAKE_AUTOMOC ON)" in the top file and replace "qt4" prefix in macros by "qt5". Added simpler macro "setup_qt()" in FindBootstrap.cmake. * Change: No longer need to include <QGtkStyle> is is choosen by default according to the current desktop environment. * Change: In <hurricane>, In HApplication, launch ExceptionWidget when a std::exception is catched instead of silently discarting it. New placer Tool: Etesian * New: <etesian> analytical placer. Encapsulate Coloquinte from Gabriel Gouvine. * New: in <documentation>, add stub demonstration ToolEngine <smurf>. Needs to be commented. Miscellaneous: * New: in <boostrap> and <unicorn>, added support for Etesian, the new analytic placer. The tool itself will be added in the next commit. * Bug: in <CellWidget>, when shifting the display buffer, we no longer can copy the buffer on itself (we should never have). Now go through a temporary one (PlaneId::AutoCopy) which is added to the DrawingPlanes. Affect "goLeft()" and "goUp()". * Bug: In <CellWidget>, remove the WA_PaintOnScreen flag/attribute. When it's on, no PaintEvent is transmitted to the CellWidget when it's the central widget of the <CellViewer> (QMainWindow). It's something I still don't understand from the doc of Qt. * Change: In <AreaCommand>, use the PlaneId enumeration instead of a anonymous numerical index. * Change: In <HApplication>, no longer catch and silently discard standartd exceptions but launch the ExceptionWidget... Suppress the now deprecated constructor with "Type" argument. * Change: In <SelectionModel>, the "reset()" method is deprecated in Qt5, instead enclose the "clear()" by a "beginResetModel()" and "endResetModel()" pair. * New: In <crlcore>, add support for ISPD05 benchmarks (in Bookshelf format). Forked from ISPD04 and not finished yet. * Change: In <Mauka>, distinguish the Action string identifier from <Etesian> * New: In <unicorn>, add entry for ISPD05 loader. Add entry for <Etesian> analytic placer.
2014-03-22 05:50:36 -05:00
{
AllianceFramework* af = AllianceFramework::get();
DbU::Unit hpitch = getHorizontalPitch();
DbU::Unit vpitch = getVerticalPitch();
cmess1 << " o Converting <" << getCell()->getName() << "> into Coloquinte." << endl;
cmess1 << ::Dots::asString(" - H-pitch" , DbU::getValueString(hpitch)) << endl;
cmess1 << ::Dots::asString(" - V-pitch" , DbU::getValueString(vpitch)) << endl;
resetPlacement();
Dots dots ( cmess2, " ", 80, 1000 );
if (not cmess2.enabled()) dots.disable();
Update to Qt 5, requires cmake 2.8.9. New placer: Etesian. Update to Qt 5: * Change: Now requires at least cmake 2.8.9. * Change: CMakeLists.txt needs small changes. Qt modules must be found one by one (Core, Gui, Widgets). Must add "set(CMAKE_AUTOMOC ON)" in the top file and replace "qt4" prefix in macros by "qt5". Added simpler macro "setup_qt()" in FindBootstrap.cmake. * Change: No longer need to include <QGtkStyle> is is choosen by default according to the current desktop environment. * Change: In <hurricane>, In HApplication, launch ExceptionWidget when a std::exception is catched instead of silently discarting it. New placer Tool: Etesian * New: <etesian> analytical placer. Encapsulate Coloquinte from Gabriel Gouvine. * New: in <documentation>, add stub demonstration ToolEngine <smurf>. Needs to be commented. Miscellaneous: * New: in <boostrap> and <unicorn>, added support for Etesian, the new analytic placer. The tool itself will be added in the next commit. * Bug: in <CellWidget>, when shifting the display buffer, we no longer can copy the buffer on itself (we should never have). Now go through a temporary one (PlaneId::AutoCopy) which is added to the DrawingPlanes. Affect "goLeft()" and "goUp()". * Bug: In <CellWidget>, remove the WA_PaintOnScreen flag/attribute. When it's on, no PaintEvent is transmitted to the CellWidget when it's the central widget of the <CellViewer> (QMainWindow). It's something I still don't understand from the doc of Qt. * Change: In <AreaCommand>, use the PlaneId enumeration instead of a anonymous numerical index. * Change: In <HApplication>, no longer catch and silently discard standartd exceptions but launch the ExceptionWidget... Suppress the now deprecated constructor with "Type" argument. * Change: In <SelectionModel>, the "reset()" method is deprecated in Qt5, instead enclose the "clear()" by a "beginResetModel()" and "endResetModel()" pair. * New: In <crlcore>, add support for ISPD05 benchmarks (in Bookshelf format). Forked from ISPD04 and not finished yet. * Change: In <Mauka>, distinguish the Action string identifier from <Etesian> * New: In <unicorn>, add entry for ISPD05 loader. Add entry for <Etesian> analytic placer.
2014-03-22 05:50:36 -05:00
size_t instancesNb = 0;
for ( Occurrence occurrence : getCell()->getLeafInstanceOccurrences() ) {
Instance* instance = static_cast<Instance*>(occurrence.getEntity());
Cell* masterCell = instance->getMasterCell();
if ( (instance->getPlacementStatus() != Instance::PlacementStatus::PLACED)
and (instance->getPlacementStatus() != Instance::PlacementStatus::FIXED )
and (masterCell->getAbutmentBox().getHeight() != getSliceHeight()) ) {
throw Error( "EtesianEngine::toColoquinte(): Cannot manage unplaced block, instance \"%s\" of \"%s\"."
, getString(instance ->getName()).c_str()
, getString(masterCell->getName()).c_str()
);
}
++instancesNb;
}
Update to Qt 5, requires cmake 2.8.9. New placer: Etesian. Update to Qt 5: * Change: Now requires at least cmake 2.8.9. * Change: CMakeLists.txt needs small changes. Qt modules must be found one by one (Core, Gui, Widgets). Must add "set(CMAKE_AUTOMOC ON)" in the top file and replace "qt4" prefix in macros by "qt5". Added simpler macro "setup_qt()" in FindBootstrap.cmake. * Change: No longer need to include <QGtkStyle> is is choosen by default according to the current desktop environment. * Change: In <hurricane>, In HApplication, launch ExceptionWidget when a std::exception is catched instead of silently discarting it. New placer Tool: Etesian * New: <etesian> analytical placer. Encapsulate Coloquinte from Gabriel Gouvine. * New: in <documentation>, add stub demonstration ToolEngine <smurf>. Needs to be commented. Miscellaneous: * New: in <boostrap> and <unicorn>, added support for Etesian, the new analytic placer. The tool itself will be added in the next commit. * Bug: in <CellWidget>, when shifting the display buffer, we no longer can copy the buffer on itself (we should never have). Now go through a temporary one (PlaneId::AutoCopy) which is added to the DrawingPlanes. Affect "goLeft()" and "goUp()". * Bug: In <CellWidget>, remove the WA_PaintOnScreen flag/attribute. When it's on, no PaintEvent is transmitted to the CellWidget when it's the central widget of the <CellViewer> (QMainWindow). It's something I still don't understand from the doc of Qt. * Change: In <AreaCommand>, use the PlaneId enumeration instead of a anonymous numerical index. * Change: In <HApplication>, no longer catch and silently discard standartd exceptions but launch the ExceptionWidget... Suppress the now deprecated constructor with "Type" argument. * Change: In <SelectionModel>, the "reset()" method is deprecated in Qt5, instead enclose the "clear()" by a "beginResetModel()" and "endResetModel()" pair. * New: In <crlcore>, add support for ISPD05 benchmarks (in Bookshelf format). Forked from ISPD04 and not finished yet. * Change: In <Mauka>, distinguish the Action string identifier from <Etesian> * New: In <unicorn>, add entry for ISPD05 loader. Add entry for <Etesian> analytic placer.
2014-03-22 05:50:36 -05:00
// Coloquinte circuit description data-structures.
Etesian integration with Chip & ClockTree plugins. * New: In Hurricane, in DebugSession, add a new method to activate the trace inconditionally with a certain level. * New: In Hurricane, in HyperNet, allow copy construction as there is no reason to disallow it and we need it now. * New: In Hurricane, in Cell::flattenNets(), add a new option to prevent the flattening of the clock net(s). For more safety perform the DeepNet creation *outside* the Collection loop. * Bug: In Hurricane, in Cell_HyperNetRootNetOccurrences, skip the DeepNets because they are the result of another flattening operation. * New: In Isobar, in PyBasicLayer, export C++ method getBlockageLayer(). * New: In Isobar, in PyRoutingGauge, export C++ method getLayerPitch(), needed by Cumulus plugins. * New: In Etesian, EtesianEngine::findYSpin() to look for the Y orientation if some cells are already placed. Typically the buffers of a clock tree. Pass the correct orienation to row_compatible_orientation(). Do not try to add feeds in the ISPD05 benchmarks. For now the benchmarks are detected through their names (unreliable). * Change: In Knik, in KnikEngine::initGlobalRouting(), allow the clock to be routed as an ordinary signal when the clock tree is not used. * New: In Kite, in BuildPowerRails, management & detection for the pre-routed clock. In KiteEngine constructor, early initialization of the blockage net to avoid later troubles in BuildPowerRails. * New: In Cumulus, in ChipPlugin, add support for Etesian plus new configuration parameter 'clockTree.placerEngine' to select between Mauka and Etesian. * New: In Cumulus, in BlockCorona, add blockages in the vertical sides in the vertical layer to prevent the router to use the vertical tracks in under the prower lines (for example, blockage in M3 with power line M5). In Cumulus, in ChipConf add attribute to access the blockage net. * New: In Cumulus, when the clock tree is disabled, do not generate the last rail around the block (the clock rail). * Bug: In Cumulus, in ChipConf use the clock pad to guess the clock signals and *not* the power pad. Add more meaningful error messages if a pad global signal is not found (implicit connexion by name). * Bug: In Cumulus, in ClockTree, compute correctly the cells Y spin, that is *from the bottom of the cell AB* (not from 'zero').
2015-02-24 06:13:17 -06:00
vector<Transformation> idsToTransf ( instancesNb );
vector<temporary_cell> instances ( instancesNb );
vector< point<int_t> > positions ( instancesNb );
Etesian integration with Chip & ClockTree plugins. * New: In Hurricane, in DebugSession, add a new method to activate the trace inconditionally with a certain level. * New: In Hurricane, in HyperNet, allow copy construction as there is no reason to disallow it and we need it now. * New: In Hurricane, in Cell::flattenNets(), add a new option to prevent the flattening of the clock net(s). For more safety perform the DeepNet creation *outside* the Collection loop. * Bug: In Hurricane, in Cell_HyperNetRootNetOccurrences, skip the DeepNets because they are the result of another flattening operation. * New: In Isobar, in PyBasicLayer, export C++ method getBlockageLayer(). * New: In Isobar, in PyRoutingGauge, export C++ method getLayerPitch(), needed by Cumulus plugins. * New: In Etesian, EtesianEngine::findYSpin() to look for the Y orientation if some cells are already placed. Typically the buffers of a clock tree. Pass the correct orienation to row_compatible_orientation(). Do not try to add feeds in the ISPD05 benchmarks. For now the benchmarks are detected through their names (unreliable). * Change: In Knik, in KnikEngine::initGlobalRouting(), allow the clock to be routed as an ordinary signal when the clock tree is not used. * New: In Kite, in BuildPowerRails, management & detection for the pre-routed clock. In KiteEngine constructor, early initialization of the blockage net to avoid later troubles in BuildPowerRails. * New: In Cumulus, in ChipPlugin, add support for Etesian plus new configuration parameter 'clockTree.placerEngine' to select between Mauka and Etesian. * New: In Cumulus, in BlockCorona, add blockages in the vertical sides in the vertical layer to prevent the router to use the vertical tracks in under the prower lines (for example, blockage in M3 with power line M5). In Cumulus, in ChipConf add attribute to access the blockage net. * New: In Cumulus, when the clock tree is disabled, do not generate the last rail around the block (the clock rail). * Bug: In Cumulus, in ChipConf use the clock pad to guess the clock signals and *not* the power pad. Add more meaningful error messages if a pad global signal is not found (implicit connexion by name). * Bug: In Cumulus, in ClockTree, compute correctly the cells Y spin, that is *from the bottom of the cell AB* (not from 'zero').
2015-02-24 06:13:17 -06:00
vector< point<bool> > orientations( instancesNb, point<bool>(true, true) );
Update to Qt 5, requires cmake 2.8.9. New placer: Etesian. Update to Qt 5: * Change: Now requires at least cmake 2.8.9. * Change: CMakeLists.txt needs small changes. Qt modules must be found one by one (Core, Gui, Widgets). Must add "set(CMAKE_AUTOMOC ON)" in the top file and replace "qt4" prefix in macros by "qt5". Added simpler macro "setup_qt()" in FindBootstrap.cmake. * Change: No longer need to include <QGtkStyle> is is choosen by default according to the current desktop environment. * Change: In <hurricane>, In HApplication, launch ExceptionWidget when a std::exception is catched instead of silently discarting it. New placer Tool: Etesian * New: <etesian> analytical placer. Encapsulate Coloquinte from Gabriel Gouvine. * New: in <documentation>, add stub demonstration ToolEngine <smurf>. Needs to be commented. Miscellaneous: * New: in <boostrap> and <unicorn>, added support for Etesian, the new analytic placer. The tool itself will be added in the next commit. * Bug: in <CellWidget>, when shifting the display buffer, we no longer can copy the buffer on itself (we should never have). Now go through a temporary one (PlaneId::AutoCopy) which is added to the DrawingPlanes. Affect "goLeft()" and "goUp()". * Bug: In <CellWidget>, remove the WA_PaintOnScreen flag/attribute. When it's on, no PaintEvent is transmitted to the CellWidget when it's the central widget of the <CellViewer> (QMainWindow). It's something I still don't understand from the doc of Qt. * Change: In <AreaCommand>, use the PlaneId enumeration instead of a anonymous numerical index. * Change: In <HApplication>, no longer catch and silently discard standartd exceptions but launch the ExceptionWidget... Suppress the now deprecated constructor with "Type" argument. * Change: In <SelectionModel>, the "reset()" method is deprecated in Qt5, instead enclose the "clear()" by a "beginResetModel()" and "endResetModel()" pair. * New: In <crlcore>, add support for ISPD05 benchmarks (in Bookshelf format). Forked from ISPD04 and not finished yet. * Change: In <Mauka>, distinguish the Action string identifier from <Etesian> * New: In <unicorn>, add entry for ISPD05 loader. Add entry for <Etesian> analytic placer.
2014-03-22 05:50:36 -05:00
cmess1 << " - Converting " << instancesNb << " instances" << endl;
cout.flush();
Box topAb = getCell()->getAbutmentBox();
UpdateSession::open();
for ( Occurrence occurrence : getCell()->getNonLeafInstanceOccurrences() )
{
Instance* instance = static_cast<Instance*>(occurrence.getEntity());
Cell* masterCell = instance->getMasterCell();
if (not masterCell->getAbutmentBox().isEmpty()
and (instance->getPlacementStatus() != Instance::PlacementStatus::PLACED)
and (instance->getPlacementStatus() != Instance::PlacementStatus::FIXED ) ) {
throw Error( "EtesianEngine::toColoquinte(): Non-leaf instance \"%s\" of \"%s\" has an abutment box but is *not* placed."
, getString(instance ->getName()).c_str()
, getString(masterCell->getName()).c_str()
);
}
Merged QuadTrees in overlayed Cells (placed together). * New: In Hurricane, in Cell & Instance, add the ability to merge the QuadTree when *second level* instances of a Cell are placed in the same space as the top Cell. This is the case of a deeply hierarchical design made of only standard cells that are to be placed in a "flat" manner. The design is uniquified then the intermediate instances models, which should be unique at that point have their QuadTree merged through a call to Instance::slaveAbutmentBox(). That method will make the model of the instance use the QuadTree of the Cell to which the instance belong. The instance model no longer posseses a dedicated QuadTree. As a corollary the abutment box of both Cell are kept identical and the Instance has it's transformation set to (0,0,ID). Remark: when we talk about "QuadTree", we mean in fact the QuadTree for the instances *and* the SliceMap (Layer+QuadTree). Consequence in Query: when going through the resulting "flattened" QuadTree we will find objects with an incomplete Path du to the fact that we didn't have to explore their Instance/Cell level to reach them. The shunted part of the Path is stored in the Go master Cell in the _shuntedPath attribute. This also affect the displayed depth of hierarchy, but not too badly. * New: In Hurricane, in Cell, new methods: - Cell::updatePlacedFlag() : set the placement flags. - Cell::isUnique() : one or less instance. - Cell::isUniquified() : is the result of an uniquification. - Cell::isUniquifyMaster() : is the reference cell of the uniquification. * Change: In Hurricane, in Cell::Uniquify(), uniquify a Cell only if it is unplaced. We do not need to duplicate placed Cells (see datapathes).
2015-06-26 11:35:11 -05:00
if ( masterCell->getAbutmentBox().isEmpty()
or ( (masterCell->getAbutmentBox().getHeight() == topAb.getHeight())
and (masterCell->getAbutmentBox().getWidth () == topAb.getWidth ()) ) ) {
// Have to check here if the model is fully placed or not.
Merged QuadTrees in overlayed Cells (placed together). * New: In Hurricane, in Cell & Instance, add the ability to merge the QuadTree when *second level* instances of a Cell are placed in the same space as the top Cell. This is the case of a deeply hierarchical design made of only standard cells that are to be placed in a "flat" manner. The design is uniquified then the intermediate instances models, which should be unique at that point have their QuadTree merged through a call to Instance::slaveAbutmentBox(). That method will make the model of the instance use the QuadTree of the Cell to which the instance belong. The instance model no longer posseses a dedicated QuadTree. As a corollary the abutment box of both Cell are kept identical and the Instance has it's transformation set to (0,0,ID). Remark: when we talk about "QuadTree", we mean in fact the QuadTree for the instances *and* the SliceMap (Layer+QuadTree). Consequence in Query: when going through the resulting "flattened" QuadTree we will find objects with an incomplete Path du to the fact that we didn't have to explore their Instance/Cell level to reach them. The shunted part of the Path is stored in the Go master Cell in the _shuntedPath attribute. This also affect the displayed depth of hierarchy, but not too badly. * New: In Hurricane, in Cell, new methods: - Cell::updatePlacedFlag() : set the placement flags. - Cell::isUnique() : one or less instance. - Cell::isUniquified() : is the result of an uniquification. - Cell::isUniquifyMaster() : is the reference cell of the uniquification. * Change: In Hurricane, in Cell::Uniquify(), uniquify a Cell only if it is unplaced. We do not need to duplicate placed Cells (see datapathes).
2015-06-26 11:35:11 -05:00
//masterCell->setAbutmentBox( topAb );
//instance->setTransformation( Transformation() ); // (0,0,ID).
//instance->setPlacementStatus( Instance::PlacementStatus::PLACED );
occurrence.makeInvalid();
Merged QuadTrees in overlayed Cells (placed together). * New: In Hurricane, in Cell & Instance, add the ability to merge the QuadTree when *second level* instances of a Cell are placed in the same space as the top Cell. This is the case of a deeply hierarchical design made of only standard cells that are to be placed in a "flat" manner. The design is uniquified then the intermediate instances models, which should be unique at that point have their QuadTree merged through a call to Instance::slaveAbutmentBox(). That method will make the model of the instance use the QuadTree of the Cell to which the instance belong. The instance model no longer posseses a dedicated QuadTree. As a corollary the abutment box of both Cell are kept identical and the Instance has it's transformation set to (0,0,ID). Remark: when we talk about "QuadTree", we mean in fact the QuadTree for the instances *and* the SliceMap (Layer+QuadTree). Consequence in Query: when going through the resulting "flattened" QuadTree we will find objects with an incomplete Path du to the fact that we didn't have to explore their Instance/Cell level to reach them. The shunted part of the Path is stored in the Go master Cell in the _shuntedPath attribute. This also affect the displayed depth of hierarchy, but not too badly. * New: In Hurricane, in Cell, new methods: - Cell::updatePlacedFlag() : set the placement flags. - Cell::isUnique() : one or less instance. - Cell::isUniquified() : is the result of an uniquification. - Cell::isUniquifyMaster() : is the reference cell of the uniquification. * Change: In Hurricane, in Cell::Uniquify(), uniquify a Cell only if it is unplaced. We do not need to duplicate placed Cells (see datapathes).
2015-06-26 11:35:11 -05:00
instance->slaveAbutmentBox();
}
}
UpdateSession::close();
Implementation of DataBase native save/restore in JSON (step 1). * New: In Hurricane, added first support for DataBase native import/export using JSON. We choose RapidJSON, in SAX mode, to manage the JSON format low level Read/Write. Thus, it's Git repository http://github.com/miloyip/rapidjson must be added under ~/coriolis-2.x/src and manually build and installed in the Coriolis installation tree (to be integrated in ccb later). Two mode are being supported: 1. Cell mode: one Cell only is saved. In that mode, Entities referred by Occurrences are "outside" the file. They are coded through their "signature" (mostly, all the values of their attributes). The ids saved in the file cannot be restored identically as we cannot predict when and in which context the Cell will be reloaded. 2. Design Blob mode: the whole design hierarchy, down and including the standard cells is saved. This way the design is completly self contained and Entities ared referred through their ids. A design blob can only be loaded immediatly after starting cgt as the DataBase *must* be empty. This way we restore the whole design hierarchy with *exactly* the same ids. Now, Hurricane object should provide a "toJson()" method for driving JSON, and be associated with a JsonObject derived class for parsing. * New: In Hurricane, ability to force the next id that will be used for a DBo (used by Design Blob Mode). * New: In Hurricane, in DataBase, added getCell() and getLibrary() functions to allow the hierarchical access of a Cell/Library in native mode (i.e. whithout the requirement of AllianceFramework). * New: In Hurricane, In CellViewer, added menu entry for Save/Load of JSON Design Blobs. Added at this level because we consider it as the "native" format of Hurricane. * New: In Unicorn, added support of import/export of JSON Cell. * Bug: In Hurricane, in Instance, when cloning an Instance or uniquifying it's master Cell, we forgot about the Occurrences (through shared pathes). When an instance is cloned the Shared pathes still points toward the original Instance. And when it's the master Cell that is uniquifyed it's the Entities pointed to that remains in the original Cell. This is a software design problem. It is difficult to define what policy to adopt when uniquifying: basically that means that one Occurence is either moved onto the clone or duplicated. Furthermore, it is not trivial to known what Occurrence is pointing on the uniquifyed/cloned item. Have to think about it a little more. * Bug: In Etesian, in EtesianEngine, build the flattened nets and their RoutingPads *after* uniquifying (through slaving bounding boxes). This way we avoid the Occurrences problem described above. * Bug: In Etesian, in EtesianEngine, invalidate the RoutingPad after processing the placement so they are put into the right quadtree. This problem is due to the fact that the RoutingPads do not belong to the Instance that they refer. And when this instance is moved around, she doesn't inform the RoutingPad that is has moved. More software architecture design to review...
2016-01-07 06:13:16 -06:00
cmess1 << " - Building RoutingPads (transhierarchical) ..." << endl;
//getCell()->flattenNets( Cell::Flags::BuildRings|Cell::Flags::NoClockFlatten );
getCell()->flattenNets( Cell::Flags::NoClockFlatten );
index_t instanceId = 0;
for ( Occurrence occurrence : getCell()->getLeafInstanceOccurrences() )
Update to Qt 5, requires cmake 2.8.9. New placer: Etesian. Update to Qt 5: * Change: Now requires at least cmake 2.8.9. * Change: CMakeLists.txt needs small changes. Qt modules must be found one by one (Core, Gui, Widgets). Must add "set(CMAKE_AUTOMOC ON)" in the top file and replace "qt4" prefix in macros by "qt5". Added simpler macro "setup_qt()" in FindBootstrap.cmake. * Change: No longer need to include <QGtkStyle> is is choosen by default according to the current desktop environment. * Change: In <hurricane>, In HApplication, launch ExceptionWidget when a std::exception is catched instead of silently discarting it. New placer Tool: Etesian * New: <etesian> analytical placer. Encapsulate Coloquinte from Gabriel Gouvine. * New: in <documentation>, add stub demonstration ToolEngine <smurf>. Needs to be commented. Miscellaneous: * New: in <boostrap> and <unicorn>, added support for Etesian, the new analytic placer. The tool itself will be added in the next commit. * Bug: in <CellWidget>, when shifting the display buffer, we no longer can copy the buffer on itself (we should never have). Now go through a temporary one (PlaneId::AutoCopy) which is added to the DrawingPlanes. Affect "goLeft()" and "goUp()". * Bug: In <CellWidget>, remove the WA_PaintOnScreen flag/attribute. When it's on, no PaintEvent is transmitted to the CellWidget when it's the central widget of the <CellViewer> (QMainWindow). It's something I still don't understand from the doc of Qt. * Change: In <AreaCommand>, use the PlaneId enumeration instead of a anonymous numerical index. * Change: In <HApplication>, no longer catch and silently discard standartd exceptions but launch the ExceptionWidget... Suppress the now deprecated constructor with "Type" argument. * Change: In <SelectionModel>, the "reset()" method is deprecated in Qt5, instead enclose the "clear()" by a "beginResetModel()" and "endResetModel()" pair. * New: In <crlcore>, add support for ISPD05 benchmarks (in Bookshelf format). Forked from ISPD04 and not finished yet. * Change: In <Mauka>, distinguish the Action string identifier from <Etesian> * New: In <unicorn>, add entry for ISPD05 loader. Add entry for <Etesian> analytic placer.
2014-03-22 05:50:36 -05:00
{
Instance* instance = static_cast<Instance*>(occurrence.getEntity());
Update to Qt 5, requires cmake 2.8.9. New placer: Etesian. Update to Qt 5: * Change: Now requires at least cmake 2.8.9. * Change: CMakeLists.txt needs small changes. Qt modules must be found one by one (Core, Gui, Widgets). Must add "set(CMAKE_AUTOMOC ON)" in the top file and replace "qt4" prefix in macros by "qt5". Added simpler macro "setup_qt()" in FindBootstrap.cmake. * Change: No longer need to include <QGtkStyle> is is choosen by default according to the current desktop environment. * Change: In <hurricane>, In HApplication, launch ExceptionWidget when a std::exception is catched instead of silently discarting it. New placer Tool: Etesian * New: <etesian> analytical placer. Encapsulate Coloquinte from Gabriel Gouvine. * New: in <documentation>, add stub demonstration ToolEngine <smurf>. Needs to be commented. Miscellaneous: * New: in <boostrap> and <unicorn>, added support for Etesian, the new analytic placer. The tool itself will be added in the next commit. * Bug: in <CellWidget>, when shifting the display buffer, we no longer can copy the buffer on itself (we should never have). Now go through a temporary one (PlaneId::AutoCopy) which is added to the DrawingPlanes. Affect "goLeft()" and "goUp()". * Bug: In <CellWidget>, remove the WA_PaintOnScreen flag/attribute. When it's on, no PaintEvent is transmitted to the CellWidget when it's the central widget of the <CellViewer> (QMainWindow). It's something I still don't understand from the doc of Qt. * Change: In <AreaCommand>, use the PlaneId enumeration instead of a anonymous numerical index. * Change: In <HApplication>, no longer catch and silently discard standartd exceptions but launch the ExceptionWidget... Suppress the now deprecated constructor with "Type" argument. * Change: In <SelectionModel>, the "reset()" method is deprecated in Qt5, instead enclose the "clear()" by a "beginResetModel()" and "endResetModel()" pair. * New: In <crlcore>, add support for ISPD05 benchmarks (in Bookshelf format). Forked from ISPD04 and not finished yet. * Change: In <Mauka>, distinguish the Action string identifier from <Etesian> * New: In <unicorn>, add entry for ISPD05 loader. Add entry for <Etesian> analytic placer.
2014-03-22 05:50:36 -05:00
Cell* masterCell = instance->getMasterCell();
string instanceName = occurrence.getCompactString();
Update to Qt 5, requires cmake 2.8.9. New placer: Etesian. Update to Qt 5: * Change: Now requires at least cmake 2.8.9. * Change: CMakeLists.txt needs small changes. Qt modules must be found one by one (Core, Gui, Widgets). Must add "set(CMAKE_AUTOMOC ON)" in the top file and replace "qt4" prefix in macros by "qt5". Added simpler macro "setup_qt()" in FindBootstrap.cmake. * Change: No longer need to include <QGtkStyle> is is choosen by default according to the current desktop environment. * Change: In <hurricane>, In HApplication, launch ExceptionWidget when a std::exception is catched instead of silently discarting it. New placer Tool: Etesian * New: <etesian> analytical placer. Encapsulate Coloquinte from Gabriel Gouvine. * New: in <documentation>, add stub demonstration ToolEngine <smurf>. Needs to be commented. Miscellaneous: * New: in <boostrap> and <unicorn>, added support for Etesian, the new analytic placer. The tool itself will be added in the next commit. * Bug: in <CellWidget>, when shifting the display buffer, we no longer can copy the buffer on itself (we should never have). Now go through a temporary one (PlaneId::AutoCopy) which is added to the DrawingPlanes. Affect "goLeft()" and "goUp()". * Bug: In <CellWidget>, remove the WA_PaintOnScreen flag/attribute. When it's on, no PaintEvent is transmitted to the CellWidget when it's the central widget of the <CellViewer> (QMainWindow). It's something I still don't understand from the doc of Qt. * Change: In <AreaCommand>, use the PlaneId enumeration instead of a anonymous numerical index. * Change: In <HApplication>, no longer catch and silently discard standartd exceptions but launch the ExceptionWidget... Suppress the now deprecated constructor with "Type" argument. * Change: In <SelectionModel>, the "reset()" method is deprecated in Qt5, instead enclose the "clear()" by a "beginResetModel()" and "endResetModel()" pair. * New: In <crlcore>, add support for ISPD05 benchmarks (in Bookshelf format). Forked from ISPD04 and not finished yet. * Change: In <Mauka>, distinguish the Action string identifier from <Etesian> * New: In <unicorn>, add entry for ISPD05 loader. Add entry for <Etesian> analytic placer.
2014-03-22 05:50:36 -05:00
// Remove the enclosing brackets...
instanceName.erase( 0, 1 );
instanceName.erase( instanceName.size()-1 );
if (CatalogExtension::isFeed(masterCell)) {
cerr << Warning("Feed instance found and skipped.") << endl;
continue;
}
Box instanceAb = masterCell->getAbutmentBox();
Update to Qt 5, requires cmake 2.8.9. New placer: Etesian. Update to Qt 5: * Change: Now requires at least cmake 2.8.9. * Change: CMakeLists.txt needs small changes. Qt modules must be found one by one (Core, Gui, Widgets). Must add "set(CMAKE_AUTOMOC ON)" in the top file and replace "qt4" prefix in macros by "qt5". Added simpler macro "setup_qt()" in FindBootstrap.cmake. * Change: No longer need to include <QGtkStyle> is is choosen by default according to the current desktop environment. * Change: In <hurricane>, In HApplication, launch ExceptionWidget when a std::exception is catched instead of silently discarting it. New placer Tool: Etesian * New: <etesian> analytical placer. Encapsulate Coloquinte from Gabriel Gouvine. * New: in <documentation>, add stub demonstration ToolEngine <smurf>. Needs to be commented. Miscellaneous: * New: in <boostrap> and <unicorn>, added support for Etesian, the new analytic placer. The tool itself will be added in the next commit. * Bug: in <CellWidget>, when shifting the display buffer, we no longer can copy the buffer on itself (we should never have). Now go through a temporary one (PlaneId::AutoCopy) which is added to the DrawingPlanes. Affect "goLeft()" and "goUp()". * Bug: In <CellWidget>, remove the WA_PaintOnScreen flag/attribute. When it's on, no PaintEvent is transmitted to the CellWidget when it's the central widget of the <CellViewer> (QMainWindow). It's something I still don't understand from the doc of Qt. * Change: In <AreaCommand>, use the PlaneId enumeration instead of a anonymous numerical index. * Change: In <HApplication>, no longer catch and silently discard standartd exceptions but launch the ExceptionWidget... Suppress the now deprecated constructor with "Type" argument. * Change: In <SelectionModel>, the "reset()" method is deprecated in Qt5, instead enclose the "clear()" by a "beginResetModel()" and "endResetModel()" pair. * New: In <crlcore>, add support for ISPD05 benchmarks (in Bookshelf format). Forked from ISPD04 and not finished yet. * Change: In <Mauka>, distinguish the Action string identifier from <Etesian> * New: In <unicorn>, add entry for ISPD05 loader. Add entry for <Etesian> analytic placer.
2014-03-22 05:50:36 -05:00
Transformation instanceTransf = instance->getTransformation();
occurrence.getPath().getTransformation().applyOn( instanceTransf );
instanceTransf.applyOn( instanceAb );
// Upper rounded
int_t xsize = (instanceAb.getWidth () + vpitch -1) / vpitch;
int_t ysize = (instanceAb.getHeight() + hpitch -1) / hpitch;
// Lower rounded
int_t xpos = instanceAb.getXMin() / vpitch;
int_t ypos = instanceAb.getYMin() / hpitch;
// if ( (ysize != 1) and not instance->isFixed() ) {
// cerr << Error( "EtesianEngine::toColoquinte(): Instance \"%s\" of \"%s\" is a block (height: %d)."
// , instanceName.c_str()
// , getString(masterCell->getName()).c_str()
// , ysize ) << endl;
// }
instances[instanceId].size = point<int_t>( xsize, ysize );
instances[instanceId].list_index = instanceId;
instances[instanceId].area = static_cast<capacity_t>(xsize) * static_cast<capacity_t>(ysize);
positions[instanceId] = point<int_t>( xpos, ypos );
if ( not instance->isFixed() and instance->isTerminal() ) {
instances[instanceId].attributes = coloquinte::XMovable
|coloquinte::YMovable
|coloquinte::XFlippable
|coloquinte::YFlippable;
} else {
instances[instanceId].attributes = 0;
}
Update to Qt 5, requires cmake 2.8.9. New placer: Etesian. Update to Qt 5: * Change: Now requires at least cmake 2.8.9. * Change: CMakeLists.txt needs small changes. Qt modules must be found one by one (Core, Gui, Widgets). Must add "set(CMAKE_AUTOMOC ON)" in the top file and replace "qt4" prefix in macros by "qt5". Added simpler macro "setup_qt()" in FindBootstrap.cmake. * Change: No longer need to include <QGtkStyle> is is choosen by default according to the current desktop environment. * Change: In <hurricane>, In HApplication, launch ExceptionWidget when a std::exception is catched instead of silently discarting it. New placer Tool: Etesian * New: <etesian> analytical placer. Encapsulate Coloquinte from Gabriel Gouvine. * New: in <documentation>, add stub demonstration ToolEngine <smurf>. Needs to be commented. Miscellaneous: * New: in <boostrap> and <unicorn>, added support for Etesian, the new analytic placer. The tool itself will be added in the next commit. * Bug: in <CellWidget>, when shifting the display buffer, we no longer can copy the buffer on itself (we should never have). Now go through a temporary one (PlaneId::AutoCopy) which is added to the DrawingPlanes. Affect "goLeft()" and "goUp()". * Bug: In <CellWidget>, remove the WA_PaintOnScreen flag/attribute. When it's on, no PaintEvent is transmitted to the CellWidget when it's the central widget of the <CellViewer> (QMainWindow). It's something I still don't understand from the doc of Qt. * Change: In <AreaCommand>, use the PlaneId enumeration instead of a anonymous numerical index. * Change: In <HApplication>, no longer catch and silently discard standartd exceptions but launch the ExceptionWidget... Suppress the now deprecated constructor with "Type" argument. * Change: In <SelectionModel>, the "reset()" method is deprecated in Qt5, instead enclose the "clear()" by a "beginResetModel()" and "endResetModel()" pair. * New: In <crlcore>, add support for ISPD05 benchmarks (in Bookshelf format). Forked from ISPD04 and not finished yet. * Change: In <Mauka>, distinguish the Action string identifier from <Etesian> * New: In <unicorn>, add entry for ISPD05 loader. Add entry for <Etesian> analytic placer.
2014-03-22 05:50:36 -05:00
_cellsToIds.insert( make_pair(instanceName,instanceId) );
_idsToInsts.push_back( instance );
++instanceId;
dots.dot();
Update to Qt 5, requires cmake 2.8.9. New placer: Etesian. Update to Qt 5: * Change: Now requires at least cmake 2.8.9. * Change: CMakeLists.txt needs small changes. Qt modules must be found one by one (Core, Gui, Widgets). Must add "set(CMAKE_AUTOMOC ON)" in the top file and replace "qt4" prefix in macros by "qt5". Added simpler macro "setup_qt()" in FindBootstrap.cmake. * Change: No longer need to include <QGtkStyle> is is choosen by default according to the current desktop environment. * Change: In <hurricane>, In HApplication, launch ExceptionWidget when a std::exception is catched instead of silently discarting it. New placer Tool: Etesian * New: <etesian> analytical placer. Encapsulate Coloquinte from Gabriel Gouvine. * New: in <documentation>, add stub demonstration ToolEngine <smurf>. Needs to be commented. Miscellaneous: * New: in <boostrap> and <unicorn>, added support for Etesian, the new analytic placer. The tool itself will be added in the next commit. * Bug: in <CellWidget>, when shifting the display buffer, we no longer can copy the buffer on itself (we should never have). Now go through a temporary one (PlaneId::AutoCopy) which is added to the DrawingPlanes. Affect "goLeft()" and "goUp()". * Bug: In <CellWidget>, remove the WA_PaintOnScreen flag/attribute. When it's on, no PaintEvent is transmitted to the CellWidget when it's the central widget of the <CellViewer> (QMainWindow). It's something I still don't understand from the doc of Qt. * Change: In <AreaCommand>, use the PlaneId enumeration instead of a anonymous numerical index. * Change: In <HApplication>, no longer catch and silently discard standartd exceptions but launch the ExceptionWidget... Suppress the now deprecated constructor with "Type" argument. * Change: In <SelectionModel>, the "reset()" method is deprecated in Qt5, instead enclose the "clear()" by a "beginResetModel()" and "endResetModel()" pair. * New: In <crlcore>, add support for ISPD05 benchmarks (in Bookshelf format). Forked from ISPD04 and not finished yet. * Change: In <Mauka>, distinguish the Action string identifier from <Etesian> * New: In <unicorn>, add entry for ISPD05 loader. Add entry for <Etesian> analytic placer.
2014-03-22 05:50:36 -05:00
}
dots.finish( Dots::Reset|Dots::FirstDot );
size_t netsNb = getCell()->getNets().getSize();
cmess1 << " - Converting " << netsNb << " nets" << endl;
vector<temporary_net> nets ( netsNb );
vector<temporary_pin> pins;
Update to Qt 5, requires cmake 2.8.9. New placer: Etesian. Update to Qt 5: * Change: Now requires at least cmake 2.8.9. * Change: CMakeLists.txt needs small changes. Qt modules must be found one by one (Core, Gui, Widgets). Must add "set(CMAKE_AUTOMOC ON)" in the top file and replace "qt4" prefix in macros by "qt5". Added simpler macro "setup_qt()" in FindBootstrap.cmake. * Change: No longer need to include <QGtkStyle> is is choosen by default according to the current desktop environment. * Change: In <hurricane>, In HApplication, launch ExceptionWidget when a std::exception is catched instead of silently discarting it. New placer Tool: Etesian * New: <etesian> analytical placer. Encapsulate Coloquinte from Gabriel Gouvine. * New: in <documentation>, add stub demonstration ToolEngine <smurf>. Needs to be commented. Miscellaneous: * New: in <boostrap> and <unicorn>, added support for Etesian, the new analytic placer. The tool itself will be added in the next commit. * Bug: in <CellWidget>, when shifting the display buffer, we no longer can copy the buffer on itself (we should never have). Now go through a temporary one (PlaneId::AutoCopy) which is added to the DrawingPlanes. Affect "goLeft()" and "goUp()". * Bug: In <CellWidget>, remove the WA_PaintOnScreen flag/attribute. When it's on, no PaintEvent is transmitted to the CellWidget when it's the central widget of the <CellViewer> (QMainWindow). It's something I still don't understand from the doc of Qt. * Change: In <AreaCommand>, use the PlaneId enumeration instead of a anonymous numerical index. * Change: In <HApplication>, no longer catch and silently discard standartd exceptions but launch the ExceptionWidget... Suppress the now deprecated constructor with "Type" argument. * Change: In <SelectionModel>, the "reset()" method is deprecated in Qt5, instead enclose the "clear()" by a "beginResetModel()" and "endResetModel()" pair. * New: In <crlcore>, add support for ISPD05 benchmarks (in Bookshelf format). Forked from ISPD04 and not finished yet. * Change: In <Mauka>, distinguish the Action string identifier from <Etesian> * New: In <unicorn>, add entry for ISPD05 loader. Add entry for <Etesian> analytic placer.
2014-03-22 05:50:36 -05:00
unsigned int netId = 0;
for ( Net* net : getCell()->getNets() )
Update to Qt 5, requires cmake 2.8.9. New placer: Etesian. Update to Qt 5: * Change: Now requires at least cmake 2.8.9. * Change: CMakeLists.txt needs small changes. Qt modules must be found one by one (Core, Gui, Widgets). Must add "set(CMAKE_AUTOMOC ON)" in the top file and replace "qt4" prefix in macros by "qt5". Added simpler macro "setup_qt()" in FindBootstrap.cmake. * Change: No longer need to include <QGtkStyle> is is choosen by default according to the current desktop environment. * Change: In <hurricane>, In HApplication, launch ExceptionWidget when a std::exception is catched instead of silently discarting it. New placer Tool: Etesian * New: <etesian> analytical placer. Encapsulate Coloquinte from Gabriel Gouvine. * New: in <documentation>, add stub demonstration ToolEngine <smurf>. Needs to be commented. Miscellaneous: * New: in <boostrap> and <unicorn>, added support for Etesian, the new analytic placer. The tool itself will be added in the next commit. * Bug: in <CellWidget>, when shifting the display buffer, we no longer can copy the buffer on itself (we should never have). Now go through a temporary one (PlaneId::AutoCopy) which is added to the DrawingPlanes. Affect "goLeft()" and "goUp()". * Bug: In <CellWidget>, remove the WA_PaintOnScreen flag/attribute. When it's on, no PaintEvent is transmitted to the CellWidget when it's the central widget of the <CellViewer> (QMainWindow). It's something I still don't understand from the doc of Qt. * Change: In <AreaCommand>, use the PlaneId enumeration instead of a anonymous numerical index. * Change: In <HApplication>, no longer catch and silently discard standartd exceptions but launch the ExceptionWidget... Suppress the now deprecated constructor with "Type" argument. * Change: In <SelectionModel>, the "reset()" method is deprecated in Qt5, instead enclose the "clear()" by a "beginResetModel()" and "endResetModel()" pair. * New: In <crlcore>, add support for ISPD05 benchmarks (in Bookshelf format). Forked from ISPD04 and not finished yet. * Change: In <Mauka>, distinguish the Action string identifier from <Etesian> * New: In <unicorn>, add entry for ISPD05 loader. Add entry for <Etesian> analytic placer.
2014-03-22 05:50:36 -05:00
{
const char* excludedType = NULL;
if (net->getType() == Net::Type::POWER ) excludedType = "POWER";
if (net->getType() == Net::Type::GROUND) excludedType = "GROUND";
if (net->getType() == Net::Type::CLOCK ) excludedType = "CLOCK";
Update to Qt 5, requires cmake 2.8.9. New placer: Etesian. Update to Qt 5: * Change: Now requires at least cmake 2.8.9. * Change: CMakeLists.txt needs small changes. Qt modules must be found one by one (Core, Gui, Widgets). Must add "set(CMAKE_AUTOMOC ON)" in the top file and replace "qt4" prefix in macros by "qt5". Added simpler macro "setup_qt()" in FindBootstrap.cmake. * Change: No longer need to include <QGtkStyle> is is choosen by default according to the current desktop environment. * Change: In <hurricane>, In HApplication, launch ExceptionWidget when a std::exception is catched instead of silently discarting it. New placer Tool: Etesian * New: <etesian> analytical placer. Encapsulate Coloquinte from Gabriel Gouvine. * New: in <documentation>, add stub demonstration ToolEngine <smurf>. Needs to be commented. Miscellaneous: * New: in <boostrap> and <unicorn>, added support for Etesian, the new analytic placer. The tool itself will be added in the next commit. * Bug: in <CellWidget>, when shifting the display buffer, we no longer can copy the buffer on itself (we should never have). Now go through a temporary one (PlaneId::AutoCopy) which is added to the DrawingPlanes. Affect "goLeft()" and "goUp()". * Bug: In <CellWidget>, remove the WA_PaintOnScreen flag/attribute. When it's on, no PaintEvent is transmitted to the CellWidget when it's the central widget of the <CellViewer> (QMainWindow). It's something I still don't understand from the doc of Qt. * Change: In <AreaCommand>, use the PlaneId enumeration instead of a anonymous numerical index. * Change: In <HApplication>, no longer catch and silently discard standartd exceptions but launch the ExceptionWidget... Suppress the now deprecated constructor with "Type" argument. * Change: In <SelectionModel>, the "reset()" method is deprecated in Qt5, instead enclose the "clear()" by a "beginResetModel()" and "endResetModel()" pair. * New: In <crlcore>, add support for ISPD05 benchmarks (in Bookshelf format). Forked from ISPD04 and not finished yet. * Change: In <Mauka>, distinguish the Action string identifier from <Etesian> * New: In <unicorn>, add entry for ISPD05 loader. Add entry for <Etesian> analytic placer.
2014-03-22 05:50:36 -05:00
if (excludedType) {
cparanoid << Warning( "%s is not a routable net (%s,excluded)."
, getString(net).c_str(), excludedType ) << endl;
Update to Qt 5, requires cmake 2.8.9. New placer: Etesian. Update to Qt 5: * Change: Now requires at least cmake 2.8.9. * Change: CMakeLists.txt needs small changes. Qt modules must be found one by one (Core, Gui, Widgets). Must add "set(CMAKE_AUTOMOC ON)" in the top file and replace "qt4" prefix in macros by "qt5". Added simpler macro "setup_qt()" in FindBootstrap.cmake. * Change: No longer need to include <QGtkStyle> is is choosen by default according to the current desktop environment. * Change: In <hurricane>, In HApplication, launch ExceptionWidget when a std::exception is catched instead of silently discarting it. New placer Tool: Etesian * New: <etesian> analytical placer. Encapsulate Coloquinte from Gabriel Gouvine. * New: in <documentation>, add stub demonstration ToolEngine <smurf>. Needs to be commented. Miscellaneous: * New: in <boostrap> and <unicorn>, added support for Etesian, the new analytic placer. The tool itself will be added in the next commit. * Bug: in <CellWidget>, when shifting the display buffer, we no longer can copy the buffer on itself (we should never have). Now go through a temporary one (PlaneId::AutoCopy) which is added to the DrawingPlanes. Affect "goLeft()" and "goUp()". * Bug: In <CellWidget>, remove the WA_PaintOnScreen flag/attribute. When it's on, no PaintEvent is transmitted to the CellWidget when it's the central widget of the <CellViewer> (QMainWindow). It's something I still don't understand from the doc of Qt. * Change: In <AreaCommand>, use the PlaneId enumeration instead of a anonymous numerical index. * Change: In <HApplication>, no longer catch and silently discard standartd exceptions but launch the ExceptionWidget... Suppress the now deprecated constructor with "Type" argument. * Change: In <SelectionModel>, the "reset()" method is deprecated in Qt5, instead enclose the "clear()" by a "beginResetModel()" and "endResetModel()" pair. * New: In <crlcore>, add support for ISPD05 benchmarks (in Bookshelf format). Forked from ISPD04 and not finished yet. * Change: In <Mauka>, distinguish the Action string identifier from <Etesian> * New: In <unicorn>, add entry for ISPD05 loader. Add entry for <Etesian> analytic placer.
2014-03-22 05:50:36 -05:00
continue;
}
if (af->isBLOCKAGE(net->getName())) continue;
Update to Qt 5, requires cmake 2.8.9. New placer: Etesian. Update to Qt 5: * Change: Now requires at least cmake 2.8.9. * Change: CMakeLists.txt needs small changes. Qt modules must be found one by one (Core, Gui, Widgets). Must add "set(CMAKE_AUTOMOC ON)" in the top file and replace "qt4" prefix in macros by "qt5". Added simpler macro "setup_qt()" in FindBootstrap.cmake. * Change: No longer need to include <QGtkStyle> is is choosen by default according to the current desktop environment. * Change: In <hurricane>, In HApplication, launch ExceptionWidget when a std::exception is catched instead of silently discarting it. New placer Tool: Etesian * New: <etesian> analytical placer. Encapsulate Coloquinte from Gabriel Gouvine. * New: in <documentation>, add stub demonstration ToolEngine <smurf>. Needs to be commented. Miscellaneous: * New: in <boostrap> and <unicorn>, added support for Etesian, the new analytic placer. The tool itself will be added in the next commit. * Bug: in <CellWidget>, when shifting the display buffer, we no longer can copy the buffer on itself (we should never have). Now go through a temporary one (PlaneId::AutoCopy) which is added to the DrawingPlanes. Affect "goLeft()" and "goUp()". * Bug: In <CellWidget>, remove the WA_PaintOnScreen flag/attribute. When it's on, no PaintEvent is transmitted to the CellWidget when it's the central widget of the <CellViewer> (QMainWindow). It's something I still don't understand from the doc of Qt. * Change: In <AreaCommand>, use the PlaneId enumeration instead of a anonymous numerical index. * Change: In <HApplication>, no longer catch and silently discard standartd exceptions but launch the ExceptionWidget... Suppress the now deprecated constructor with "Type" argument. * Change: In <SelectionModel>, the "reset()" method is deprecated in Qt5, instead enclose the "clear()" by a "beginResetModel()" and "endResetModel()" pair. * New: In <crlcore>, add support for ISPD05 benchmarks (in Bookshelf format). Forked from ISPD04 and not finished yet. * Change: In <Mauka>, distinguish the Action string identifier from <Etesian> * New: In <unicorn>, add entry for ISPD05 loader. Add entry for <Etesian> analytic placer.
2014-03-22 05:50:36 -05:00
dots.dot();
nets[netId] = temporary_net( netId, 1 );
for ( RoutingPad* rp : net->getRoutingPads() ) {
string insName = extractInstanceName( rp );
Point offset = extractRpOffset ( rp );
int_t xpin = offset.getX() / vpitch;
int_t ypin = offset.getY() / hpitch;
Update to Qt 5, requires cmake 2.8.9. New placer: Etesian. Update to Qt 5: * Change: Now requires at least cmake 2.8.9. * Change: CMakeLists.txt needs small changes. Qt modules must be found one by one (Core, Gui, Widgets). Must add "set(CMAKE_AUTOMOC ON)" in the top file and replace "qt4" prefix in macros by "qt5". Added simpler macro "setup_qt()" in FindBootstrap.cmake. * Change: No longer need to include <QGtkStyle> is is choosen by default according to the current desktop environment. * Change: In <hurricane>, In HApplication, launch ExceptionWidget when a std::exception is catched instead of silently discarting it. New placer Tool: Etesian * New: <etesian> analytical placer. Encapsulate Coloquinte from Gabriel Gouvine. * New: in <documentation>, add stub demonstration ToolEngine <smurf>. Needs to be commented. Miscellaneous: * New: in <boostrap> and <unicorn>, added support for Etesian, the new analytic placer. The tool itself will be added in the next commit. * Bug: in <CellWidget>, when shifting the display buffer, we no longer can copy the buffer on itself (we should never have). Now go through a temporary one (PlaneId::AutoCopy) which is added to the DrawingPlanes. Affect "goLeft()" and "goUp()". * Bug: In <CellWidget>, remove the WA_PaintOnScreen flag/attribute. When it's on, no PaintEvent is transmitted to the CellWidget when it's the central widget of the <CellViewer> (QMainWindow). It's something I still don't understand from the doc of Qt. * Change: In <AreaCommand>, use the PlaneId enumeration instead of a anonymous numerical index. * Change: In <HApplication>, no longer catch and silently discard standartd exceptions but launch the ExceptionWidget... Suppress the now deprecated constructor with "Type" argument. * Change: In <SelectionModel>, the "reset()" method is deprecated in Qt5, instead enclose the "clear()" by a "beginResetModel()" and "endResetModel()" pair. * New: In <crlcore>, add support for ISPD05 benchmarks (in Bookshelf format). Forked from ISPD04 and not finished yet. * Change: In <Mauka>, distinguish the Action string identifier from <Etesian> * New: In <unicorn>, add entry for ISPD05 loader. Add entry for <Etesian> analytic placer.
2014-03-22 05:50:36 -05:00
auto iid = _cellsToIds.find( insName );
if (iid == _cellsToIds.end() ) {
cerr << Error( "Unable to lookup instance <%s>.", insName.c_str() ) << endl;
} else {
pins.push_back( temporary_pin( point<int_t>(xpin,ypin), (*iid).second, netId ) );
Update to Qt 5, requires cmake 2.8.9. New placer: Etesian. Update to Qt 5: * Change: Now requires at least cmake 2.8.9. * Change: CMakeLists.txt needs small changes. Qt modules must be found one by one (Core, Gui, Widgets). Must add "set(CMAKE_AUTOMOC ON)" in the top file and replace "qt4" prefix in macros by "qt5". Added simpler macro "setup_qt()" in FindBootstrap.cmake. * Change: No longer need to include <QGtkStyle> is is choosen by default according to the current desktop environment. * Change: In <hurricane>, In HApplication, launch ExceptionWidget when a std::exception is catched instead of silently discarting it. New placer Tool: Etesian * New: <etesian> analytical placer. Encapsulate Coloquinte from Gabriel Gouvine. * New: in <documentation>, add stub demonstration ToolEngine <smurf>. Needs to be commented. Miscellaneous: * New: in <boostrap> and <unicorn>, added support for Etesian, the new analytic placer. The tool itself will be added in the next commit. * Bug: in <CellWidget>, when shifting the display buffer, we no longer can copy the buffer on itself (we should never have). Now go through a temporary one (PlaneId::AutoCopy) which is added to the DrawingPlanes. Affect "goLeft()" and "goUp()". * Bug: In <CellWidget>, remove the WA_PaintOnScreen flag/attribute. When it's on, no PaintEvent is transmitted to the CellWidget when it's the central widget of the <CellViewer> (QMainWindow). It's something I still don't understand from the doc of Qt. * Change: In <AreaCommand>, use the PlaneId enumeration instead of a anonymous numerical index. * Change: In <HApplication>, no longer catch and silently discard standartd exceptions but launch the ExceptionWidget... Suppress the now deprecated constructor with "Type" argument. * Change: In <SelectionModel>, the "reset()" method is deprecated in Qt5, instead enclose the "clear()" by a "beginResetModel()" and "endResetModel()" pair. * New: In <crlcore>, add support for ISPD05 benchmarks (in Bookshelf format). Forked from ISPD04 and not finished yet. * Change: In <Mauka>, distinguish the Action string identifier from <Etesian> * New: In <unicorn>, add entry for ISPD05 loader. Add entry for <Etesian> analytic placer.
2014-03-22 05:50:36 -05:00
}
}
netId++;
}
dots.finish( Dots::Reset );
Update to Qt 5, requires cmake 2.8.9. New placer: Etesian. Update to Qt 5: * Change: Now requires at least cmake 2.8.9. * Change: CMakeLists.txt needs small changes. Qt modules must be found one by one (Core, Gui, Widgets). Must add "set(CMAKE_AUTOMOC ON)" in the top file and replace "qt4" prefix in macros by "qt5". Added simpler macro "setup_qt()" in FindBootstrap.cmake. * Change: No longer need to include <QGtkStyle> is is choosen by default according to the current desktop environment. * Change: In <hurricane>, In HApplication, launch ExceptionWidget when a std::exception is catched instead of silently discarting it. New placer Tool: Etesian * New: <etesian> analytical placer. Encapsulate Coloquinte from Gabriel Gouvine. * New: in <documentation>, add stub demonstration ToolEngine <smurf>. Needs to be commented. Miscellaneous: * New: in <boostrap> and <unicorn>, added support for Etesian, the new analytic placer. The tool itself will be added in the next commit. * Bug: in <CellWidget>, when shifting the display buffer, we no longer can copy the buffer on itself (we should never have). Now go through a temporary one (PlaneId::AutoCopy) which is added to the DrawingPlanes. Affect "goLeft()" and "goUp()". * Bug: In <CellWidget>, remove the WA_PaintOnScreen flag/attribute. When it's on, no PaintEvent is transmitted to the CellWidget when it's the central widget of the <CellViewer> (QMainWindow). It's something I still don't understand from the doc of Qt. * Change: In <AreaCommand>, use the PlaneId enumeration instead of a anonymous numerical index. * Change: In <HApplication>, no longer catch and silently discard standartd exceptions but launch the ExceptionWidget... Suppress the now deprecated constructor with "Type" argument. * Change: In <SelectionModel>, the "reset()" method is deprecated in Qt5, instead enclose the "clear()" by a "beginResetModel()" and "endResetModel()" pair. * New: In <crlcore>, add support for ISPD05 benchmarks (in Bookshelf format). Forked from ISPD04 and not finished yet. * Change: In <Mauka>, distinguish the Action string identifier from <Etesian> * New: In <unicorn>, add entry for ISPD05 loader. Add entry for <Etesian> analytic placer.
2014-03-22 05:50:36 -05:00
_surface = box<int_t>( (int_t)(getCell()->getAbutmentBox().getXMin() / vpitch)
, (int_t)(getCell()->getAbutmentBox().getXMax() / vpitch)
, (int_t)(getCell()->getAbutmentBox().getYMin() / hpitch)
, (int_t)(getCell()->getAbutmentBox().getYMax() / hpitch)
);
_circuit = netlist( instances, nets, pins );
_circuit.selfcheck();
_placementLB.positions_ = positions;
_placementLB.orientations_ = orientations;
_placementUB = _placementLB;
//cerr << "Coloquinte cell height: " << _circuit.get_cell(0).size.y << endl;
}
Update to Qt 5, requires cmake 2.8.9. New placer: Etesian. Update to Qt 5: * Change: Now requires at least cmake 2.8.9. * Change: CMakeLists.txt needs small changes. Qt modules must be found one by one (Core, Gui, Widgets). Must add "set(CMAKE_AUTOMOC ON)" in the top file and replace "qt4" prefix in macros by "qt5". Added simpler macro "setup_qt()" in FindBootstrap.cmake. * Change: No longer need to include <QGtkStyle> is is choosen by default according to the current desktop environment. * Change: In <hurricane>, In HApplication, launch ExceptionWidget when a std::exception is catched instead of silently discarting it. New placer Tool: Etesian * New: <etesian> analytical placer. Encapsulate Coloquinte from Gabriel Gouvine. * New: in <documentation>, add stub demonstration ToolEngine <smurf>. Needs to be commented. Miscellaneous: * New: in <boostrap> and <unicorn>, added support for Etesian, the new analytic placer. The tool itself will be added in the next commit. * Bug: in <CellWidget>, when shifting the display buffer, we no longer can copy the buffer on itself (we should never have). Now go through a temporary one (PlaneId::AutoCopy) which is added to the DrawingPlanes. Affect "goLeft()" and "goUp()". * Bug: In <CellWidget>, remove the WA_PaintOnScreen flag/attribute. When it's on, no PaintEvent is transmitted to the CellWidget when it's the central widget of the <CellViewer> (QMainWindow). It's something I still don't understand from the doc of Qt. * Change: In <AreaCommand>, use the PlaneId enumeration instead of a anonymous numerical index. * Change: In <HApplication>, no longer catch and silently discard standartd exceptions but launch the ExceptionWidget... Suppress the now deprecated constructor with "Type" argument. * Change: In <SelectionModel>, the "reset()" method is deprecated in Qt5, instead enclose the "clear()" by a "beginResetModel()" and "endResetModel()" pair. * New: In <crlcore>, add support for ISPD05 benchmarks (in Bookshelf format). Forked from ISPD04 and not finished yet. * Change: In <Mauka>, distinguish the Action string identifier from <Etesian> * New: In <unicorn>, add entry for ISPD05 loader. Add entry for <Etesian> analytic placer.
2014-03-22 05:50:36 -05:00
void EtesianEngine::preplace ()
{
using namespace coloquinte::gp;
// Perform a very quick legalization pass
cmess2 << " o Simple legalization." << endl;
auto first_legalizer = region_distribution::uniform_density_distribution(_surface, _circuit, _placementLB);
first_legalizer.selfcheck();
get_rough_legalization( _circuit, _placementUB, first_legalizer);
_placementLB = _placementUB;
// Early topology-independent solution with a star model + negligible pulling forces to avoid dumb solutions
// Spreads well to help subsequent optimization passes
cmess2 << " o Star (*) Optimization." << endl;
auto solv = get_star_linear_system( _circuit, _placementLB, 1.0, 0, 10) // Limit the number of pins: don't want big awful nets with high weight
+ get_pulling_forces( _circuit, _placementUB, 1000000.0);
solve_linear_system( _circuit, _placementLB, solv, 200 );
_progressReport2(" [---]" );
}
void EtesianEngine::roughLegalize ( float minDisruption, unsigned options )
{
using namespace coloquinte::gp;
// Create a legalizer and bipartition it until we have sufficient precision
auto legalizer = (options & ForceUniformDensity) != 0 ?
region_distribution::uniform_density_distribution (_surface, _circuit, _placementLB, _densityLimits)
: region_distribution::full_density_distribution (_surface, _circuit, _placementLB, _densityLimits);
while(legalizer.region_dimensions().x > 2*legalizer.region_dimensions().y)
legalizer.x_bipartition();
while(2*legalizer.region_dimensions().x < legalizer.region_dimensions().y)
legalizer.y_bipartition();
while( std::max(legalizer.region_dimensions().x, legalizer.region_dimensions().y)*4 > minDisruption ) {
legalizer.x_bipartition();
legalizer.y_bipartition();
legalizer.redo_line_partitions();
legalizer.redo_diagonal_bipartitions();
legalizer.redo_line_partitions();
legalizer.redo_diagonal_bipartitions();
legalizer.selfcheck();
}
// Keep the orientation between LB and UB
_placementUB = _placementLB;
// Update UB
get_rough_legalization( _circuit, _placementUB, legalizer );
}
void EtesianEngine::feedRoutingBack ()
{
using namespace Katabatic;
2015-04-20 09:50:30 -05:00
using namespace Kite;
/*
* If routing information is present, use it to
* * artificially expand the areas given to coloquinte
* * add placement dentity constraints
*/
DbU::Unit hpitch = getHorizontalPitch();
DbU::Unit vpitch = getVerticalPitch();
const float densityThreshold = 0.9;
2015-04-20 09:50:30 -05:00
KiteEngine* routingEngine = KiteEngine::get( getCell() );
if(routingEngine == NULL)
throw Error("No routing information was found when performing routing-driven placement\n");
GCellGrid * grid = routingEngine->getGCellGrid();
// Get information about the GCells
// Create different densities
_densityLimits.clear();
for(GCell* gc : grid->getGCells()){
float density = gc->getMaxHVDensity();
if(density >= densityThreshold){
coloquinte::density_limit cur;
cur.box_ = coloquinte::box<int_t>(
gc->getX() / vpitch,
gc->getXMax() / vpitch,
gc->getY() / hpitch,
gc->getYMax() / hpitch
);
cur.density_ = densityThreshold/density;
_densityLimits.push_back(cur);
}
}
// TODO: Careful to keep the densities high enough
// Will just fail later if the densities are too high
// Expand areas: TODO
}
void EtesianEngine::globalPlace ( float initPenalty
, float minDisruption
, float targetImprovement
, float minInc
, float maxInc
, unsigned options )
{
using namespace coloquinte::gp;
float_t penaltyIncrease = minInc;
float_t linearDisruption = get_mean_linear_disruption(_circuit, _placementLB, _placementUB);
float_t pullingForce = initPenalty;
float_t upperWL = static_cast<float_t>(get_HPWL_wirelength(_circuit, _placementUB)),
lowerWL = static_cast<float_t>(get_HPWL_wirelength(_circuit, _placementLB));
float_t prevOptRatio = lowerWL / upperWL;
index_t i=0;
do{
roughLegalize(minDisruption, options);
if(options & UpdateUB)
_updatePlacement( _placementUB );
ostringstream label;
label.str("");
label << " [" << setw(3) << setfill('0') << i << "] Bipart.";
_progressReport1(label.str() );
upperWL = static_cast<float_t>(get_HPWL_wirelength(_circuit, _placementUB));
//float_t legRatio = lowerWL / upperWL;
// Get the system to optimize (tolerance, maximum and minimum pin counts)
// and the pulling forces (threshold distance)
auto opt_problem = (options & SteinerModel) ?
get_RSMT_linear_system ( _circuit, _placementLB, minDisruption, 2, 100000 )
: get_HPWLF_linear_system ( _circuit, _placementLB, minDisruption, 2, 100000 );
auto solv = opt_problem
+ get_linear_pulling_forces( _circuit, _placementUB, _placementLB, pullingForce, 2.0f * linearDisruption);
solve_linear_system( _circuit, _placementLB, solv, 200 ); // 200 iterations
_progressReport2(" Linear." );
if(options & UpdateLB)
_updatePlacement( _placementLB );
// Optimize orientation sometimes
if (i%5 == 0) {
optimize_exact_orientations( _circuit, _placementLB );
_progressReport2(" Orient." );
}
lowerWL = static_cast<float_t>(get_HPWL_wirelength(_circuit, _placementLB));
float_t optRatio = lowerWL / upperWL;
/*
* Schedule the penalty during global placement to achieve uniform improvement
*
* Currently, the metric considered is the ratio optimized HPWL/legalized HPWL
* Other ones, like the disruption itself, may be considered
*/
penaltyIncrease = std::min(maxInc, std::max(minInc,
penaltyIncrease * std::sqrt( targetImprovement / (optRatio - prevOptRatio) )
) );
cparanoid << " L/U ratio: " << 100*optRatio << "% (previous: " << 100*prevOptRatio << "%)\n"
<< " Pulling force: " << pullingForce << " Increase: " << penaltyIncrease << endl;
pullingForce += penaltyIncrease;
prevOptRatio = optRatio;
linearDisruption = get_mean_linear_disruption(_circuit, _placementLB, _placementUB);
++i;
// First way to exit the loop: UB and LB difference is <10%
// Second way to exit the loop: the legalization is close enough to the previous result
} while (linearDisruption > minDisruption and prevOptRatio <= 0.9);
_updatePlacement( _placementUB );
}
void EtesianEngine::detailedPlace ( int iterations, int effort, unsigned options )
{
using namespace coloquinte::gp;
using namespace coloquinte::dp;
int_t sliceHeight = getSliceHeight() / getHorizontalPitch();
roughLegalize(sliceHeight, options);
for ( int i=0; i<iterations; ++i ){
ostringstream label;
label.str("");
label << " [" << setw(3) << setfill('0') << i << "]";
optimize_x_orientations( _circuit, _placementUB ); // Don't disrupt VDD/VSS connections in a row
_progressReport1(label.str()+" Oriented ......." );
if(options & UpdateDetailed)
_updatePlacement( _placementUB );
auto legalizer = legalize( _circuit, _placementUB, _surface, sliceHeight );
coloquinte::dp::get_result( _circuit, legalizer, _placementUB );
_progressReport1(" Legalized ......" );
if(options & UpdateDetailed)
_updatePlacement( _placementUB );
row_compatible_orientation( _circuit, legalizer, true );
swaps_global_HPWL( _circuit, legalizer, 3, 4 );
coloquinte::dp::get_result( _circuit, legalizer, _placementUB );
_progressReport1(" Global Swaps ..." );
if(options & UpdateDetailed)
_updatePlacement( _placementUB );
if(options & SteinerModel)
OSRP_noncvx_RSMT( _circuit, legalizer );
else
OSRP_convex_HPWL( _circuit, legalizer );
coloquinte::dp::get_result( _circuit, legalizer, _placementUB );
_progressReport1(" Row Optimization" );
if(options & UpdateDetailed)
_updatePlacement( _placementUB );
if(options & SteinerModel)
swaps_row_noncvx_RSMT( _circuit, legalizer, effort+2 );
else
swaps_row_convex_HPWL( _circuit, legalizer, effort+2 );
coloquinte::dp::get_result( _circuit, legalizer, _placementUB );
_progressReport1(" Local Swaps ...." );
if(options & UpdateDetailed)
_updatePlacement( _placementUB );
if (i == iterations-1) {
//swaps_row_convex_RSMT( _circuit, legalizer, 4 );
row_compatible_orientation( _circuit, legalizer, true );
coloquinte::dp::get_result( _circuit, legalizer, _placementUB );
verify_placement_legality( _circuit, _placementUB, _surface );
_progressReport1(" Final Legalize ." );
}
}
2015-04-20 09:50:30 -05:00
_placementLB = _placementUB; // In case we run other passes
_updatePlacement( _placementUB );
}
void EtesianEngine::place ()
{
if(getCell()->isPlaced()){
cmess2 << Warning("The cell is already placed; returning") << std::endl;
return;
}
getCell()->uniquify();
getConfiguration()->print( getCell() );
if (getCell()->getAbutmentBox().isEmpty()) setDefaultAb();
findYSpin();
toColoquinte();
Effort placementEffort = getPlaceEffort();
GraphicUpdate placementUpdate = getUpdateConf();
Density densityConf = getSpreadingConf();
bool routingDriven = getRoutingDriven();
startMeasures();
double sliceHeight = getSliceHeight() / getHorizontalPitch();
cmess1 << " o Running Coloquinte." << endl;
cmess2 << " - Computing initial placement..." << endl;
cmess2 << right;
preplace();
float_t minPenaltyIncrease, maxPenaltyIncrease, targetImprovement;
int detailedIterations, detailedEffort;
unsigned globalOptions=0, detailedOptions=0;
if(placementUpdate == UpdateAll){
globalOptions |= (UpdateUB | UpdateLB);
detailedOptions |= UpdateDetailed;
}
else if(placementUpdate == LowerBound){
globalOptions |= UpdateLB;
}
if(densityConf == ForceUniform)
globalOptions |= ForceUniformDensity;
if(placementEffort == Fast){
minPenaltyIncrease = 0.005f;
maxPenaltyIncrease = 0.08f;
targetImprovement = 0.05f; // 5/100 per iteration
detailedIterations = 1;
detailedEffort = 0;
}
else if(placementEffort == Standard){
minPenaltyIncrease = 0.001f;
maxPenaltyIncrease = 0.04f;
targetImprovement = 0.02f; // 2/100 per iteration
detailedIterations = 2;
detailedEffort = 1;
}
else if(placementEffort == High){
minPenaltyIncrease = 0.0005f;
maxPenaltyIncrease = 0.02f;
targetImprovement = 0.01f; // 1/100 per iteration
detailedIterations = 4;
detailedEffort = 2;
}
else{
minPenaltyIncrease = 0.0002f;
maxPenaltyIncrease = 0.01f;
targetImprovement = 0.005f; // 5/1000 per iteration
detailedIterations = 7;
detailedEffort = 3;
}
cmess1 << " o Global placement." << endl;
globalPlace(minPenaltyIncrease, sliceHeight, targetImprovement, minPenaltyIncrease, maxPenaltyIncrease, globalOptions);
cmess1 << " o Detailed Placement." << endl;
detailedPlace(detailedIterations, detailedEffort, detailedOptions);
if(routingDriven){
bool success = false;
int routingDrivenIteration = 0;
using namespace Kite;
while(true){
cmess2 << "Routing-driven placement iteration " << routingDrivenIteration << endl;
KiteEngine* kiteE = KiteEngine::create(_cell);
kiteE->runGlobalRouter(0);
kiteE->loadGlobalRouting(Katabatic::EngineLoadGrByNet);
kiteE->balanceGlobalDensity();
kiteE->layerAssign(Katabatic::EngineNoNetLayerAssign);
kiteE->runNegociate();
success = kiteE->getToolSuccess();
feedRoutingBack();
kiteE->destroy();
KiteEngine::wipeoutRouting(_cell);
if(success){
cmess2 << "The design is routable; exiting" << endl;
break;
}
detailedPlace(detailedIterations, detailedEffort, detailedOptions);
2015-04-23 08:16:53 -05:00
}
2015-04-20 09:50:30 -05:00
}
cmess2 << " o Adding feed cells." << endl;
addFeeds();
cmess1 << " o Placement finished." << endl;
stopMeasures();
Apaired segments building for symmetric routing (step 1). * Change: In Hurricane::BaseFlags, store flags in uint64_t instead of unsigned int because we start to need more than 32 different flags in some tools. * New: In ::getString() & ::getRecord() templates, add support for std::array<>. * Change: In CRL::ToolEngine, add support for timer (time & memory measurements) displaced from Katabatic. This way all ToolEngine can use this feature. The _postCreate() method display the memory just after ToolEngine allocation. * Change: In Etesian::EtesianEngine, make use of the ToolEngine builtin timer (remove the local one). Forgot to call the base class _postCreate() and _preDestroy(). * Change: In Anabatic::AnabaticEngine, make use of the ToolEngine builtin timer (remove the local one). * New: In Anabatic, new AutoSegments_Connecteds() collection. This Collection allows a deterministic walkthough *all* the AutoSegments connected either to source or target of one AutoSegment. * New: In Anabatic::AutoContactTerminal::isEndPoint() to check if an AutoContactTerminal is the *only one* anchored on a RoutingPad, thus being a true "end point" and not a kind of feed-through. * New: In Katana::KatanaEngine, added support for symmetric nets. Created new class DataSymmetric to store symmetric information of a net (mainly the paired AutoSegments). Added KatanaEngine::runSymmetricRouter(), for now only build the DataSymmetric informations. More to come... * Change: In Katana::GraphicKatanaEngine::_runTest(), now perform symmetric information building the non-symmetric routing.
2017-03-12 13:34:12 -05:00
printMeasures();
cmess1 << ::Dots::asString
( " - HPWL", DbU::getValueString( (DbU::Unit)coloquinte::gp::get_HPWL_wirelength(_circuit,_placementUB )*getVerticalPitch() ) ) << endl;
cmess1 << ::Dots::asString
( " - RMST", DbU::getValueString( (DbU::Unit)coloquinte::gp::get_RSMT_wirelength(_circuit,_placementUB )*getVerticalPitch() ) ) << endl;
_placed = true;
Implementation of DataBase native save/restore in JSON (step 1). * New: In Hurricane, added first support for DataBase native import/export using JSON. We choose RapidJSON, in SAX mode, to manage the JSON format low level Read/Write. Thus, it's Git repository http://github.com/miloyip/rapidjson must be added under ~/coriolis-2.x/src and manually build and installed in the Coriolis installation tree (to be integrated in ccb later). Two mode are being supported: 1. Cell mode: one Cell only is saved. In that mode, Entities referred by Occurrences are "outside" the file. They are coded through their "signature" (mostly, all the values of their attributes). The ids saved in the file cannot be restored identically as we cannot predict when and in which context the Cell will be reloaded. 2. Design Blob mode: the whole design hierarchy, down and including the standard cells is saved. This way the design is completly self contained and Entities ared referred through their ids. A design blob can only be loaded immediatly after starting cgt as the DataBase *must* be empty. This way we restore the whole design hierarchy with *exactly* the same ids. Now, Hurricane object should provide a "toJson()" method for driving JSON, and be associated with a JsonObject derived class for parsing. * New: In Hurricane, ability to force the next id that will be used for a DBo (used by Design Blob Mode). * New: In Hurricane, in DataBase, added getCell() and getLibrary() functions to allow the hierarchical access of a Cell/Library in native mode (i.e. whithout the requirement of AllianceFramework). * New: In Hurricane, In CellViewer, added menu entry for Save/Load of JSON Design Blobs. Added at this level because we consider it as the "native" format of Hurricane. * New: In Unicorn, added support of import/export of JSON Cell. * Bug: In Hurricane, in Instance, when cloning an Instance or uniquifying it's master Cell, we forgot about the Occurrences (through shared pathes). When an instance is cloned the Shared pathes still points toward the original Instance. And when it's the master Cell that is uniquifyed it's the Entities pointed to that remains in the original Cell. This is a software design problem. It is difficult to define what policy to adopt when uniquifying: basically that means that one Occurence is either moved onto the clone or duplicated. Furthermore, it is not trivial to known what Occurrence is pointing on the uniquifyed/cloned item. Have to think about it a little more. * Bug: In Etesian, in EtesianEngine, build the flattened nets and their RoutingPads *after* uniquifying (through slaving bounding boxes). This way we avoid the Occurrences problem described above. * Bug: In Etesian, in EtesianEngine, invalidate the RoutingPad after processing the placement so they are put into the right quadtree. This problem is due to the fact that the RoutingPads do not belong to the Instance that they refer. And when this instance is moved around, she doesn't inform the RoutingPad that is has moved. More software architecture design to review...
2016-01-07 06:13:16 -06:00
UpdateSession::open();
for ( Net* net : getCell()->getNets() ) {
for ( RoutingPad* rp : net->getComponents().getSubSet<RoutingPad*>() ) {
rp->invalidate();
}
}
UpdateSession::close();
getCell()->setFlags( Cell::Flags::Placed );
Update to Qt 5, requires cmake 2.8.9. New placer: Etesian. Update to Qt 5: * Change: Now requires at least cmake 2.8.9. * Change: CMakeLists.txt needs small changes. Qt modules must be found one by one (Core, Gui, Widgets). Must add "set(CMAKE_AUTOMOC ON)" in the top file and replace "qt4" prefix in macros by "qt5". Added simpler macro "setup_qt()" in FindBootstrap.cmake. * Change: No longer need to include <QGtkStyle> is is choosen by default according to the current desktop environment. * Change: In <hurricane>, In HApplication, launch ExceptionWidget when a std::exception is catched instead of silently discarting it. New placer Tool: Etesian * New: <etesian> analytical placer. Encapsulate Coloquinte from Gabriel Gouvine. * New: in <documentation>, add stub demonstration ToolEngine <smurf>. Needs to be commented. Miscellaneous: * New: in <boostrap> and <unicorn>, added support for Etesian, the new analytic placer. The tool itself will be added in the next commit. * Bug: in <CellWidget>, when shifting the display buffer, we no longer can copy the buffer on itself (we should never have). Now go through a temporary one (PlaneId::AutoCopy) which is added to the DrawingPlanes. Affect "goLeft()" and "goUp()". * Bug: In <CellWidget>, remove the WA_PaintOnScreen flag/attribute. When it's on, no PaintEvent is transmitted to the CellWidget when it's the central widget of the <CellViewer> (QMainWindow). It's something I still don't understand from the doc of Qt. * Change: In <AreaCommand>, use the PlaneId enumeration instead of a anonymous numerical index. * Change: In <HApplication>, no longer catch and silently discard standartd exceptions but launch the ExceptionWidget... Suppress the now deprecated constructor with "Type" argument. * Change: In <SelectionModel>, the "reset()" method is deprecated in Qt5, instead enclose the "clear()" by a "beginResetModel()" and "endResetModel()" pair. * New: In <crlcore>, add support for ISPD05 benchmarks (in Bookshelf format). Forked from ISPD04 and not finished yet. * Change: In <Mauka>, distinguish the Action string identifier from <Etesian> * New: In <unicorn>, add entry for ISPD05 loader. Add entry for <Etesian> analytic placer.
2014-03-22 05:50:36 -05:00
}
void EtesianEngine::_progressReport1 ( string label ) const
{
size_t w = label.size();
string indent ( w, ' ' );
if (not w) {
label = string( 5, ' ' );
indent = label;
}
Apaired segments building for symmetric routing (step 1). * Change: In Hurricane::BaseFlags, store flags in uint64_t instead of unsigned int because we start to need more than 32 different flags in some tools. * New: In ::getString() & ::getRecord() templates, add support for std::array<>. * Change: In CRL::ToolEngine, add support for timer (time & memory measurements) displaced from Katabatic. This way all ToolEngine can use this feature. The _postCreate() method display the memory just after ToolEngine allocation. * Change: In Etesian::EtesianEngine, make use of the ToolEngine builtin timer (remove the local one). Forgot to call the base class _postCreate() and _preDestroy(). * Change: In Anabatic::AnabaticEngine, make use of the ToolEngine builtin timer (remove the local one). * New: In Anabatic, new AutoSegments_Connecteds() collection. This Collection allows a deterministic walkthough *all* the AutoSegments connected either to source or target of one AutoSegment. * New: In Anabatic::AutoContactTerminal::isEndPoint() to check if an AutoContactTerminal is the *only one* anchored on a RoutingPad, thus being a true "end point" and not a kind of feed-through. * New: In Katana::KatanaEngine, added support for symmetric nets. Created new class DataSymmetric to store symmetric information of a net (mainly the paired AutoSegments). Added KatanaEngine::runSymmetricRouter(), for now only build the DataSymmetric informations. More to come... * Change: In Katana::GraphicKatanaEngine::_runTest(), now perform symmetric information building the non-symmetric routing.
2017-03-12 13:34:12 -05:00
cmess2 << label
<< " HPWL:" << setw(11) << coloquinte::gp::get_HPWL_wirelength( _circuit, _placementUB )
<< " RMST:" << setw(11) << coloquinte::gp::get_RSMT_wirelength( _circuit, _placementUB )
<< endl;
cparanoid << indent
<< " L-Dsrpt:" << setw(8) << coloquinte::gp::get_mean_linear_disruption ( _circuit, _placementLB, _placementUB )
<< " Q-Dsrpt:" << setw(8) << coloquinte::gp::get_mean_quadratic_disruption( _circuit, _placementLB, _placementUB )
<< endl;
}
void EtesianEngine::_progressReport2 ( string label ) const
Update to Qt 5, requires cmake 2.8.9. New placer: Etesian. Update to Qt 5: * Change: Now requires at least cmake 2.8.9. * Change: CMakeLists.txt needs small changes. Qt modules must be found one by one (Core, Gui, Widgets). Must add "set(CMAKE_AUTOMOC ON)" in the top file and replace "qt4" prefix in macros by "qt5". Added simpler macro "setup_qt()" in FindBootstrap.cmake. * Change: No longer need to include <QGtkStyle> is is choosen by default according to the current desktop environment. * Change: In <hurricane>, In HApplication, launch ExceptionWidget when a std::exception is catched instead of silently discarting it. New placer Tool: Etesian * New: <etesian> analytical placer. Encapsulate Coloquinte from Gabriel Gouvine. * New: in <documentation>, add stub demonstration ToolEngine <smurf>. Needs to be commented. Miscellaneous: * New: in <boostrap> and <unicorn>, added support for Etesian, the new analytic placer. The tool itself will be added in the next commit. * Bug: in <CellWidget>, when shifting the display buffer, we no longer can copy the buffer on itself (we should never have). Now go through a temporary one (PlaneId::AutoCopy) which is added to the DrawingPlanes. Affect "goLeft()" and "goUp()". * Bug: In <CellWidget>, remove the WA_PaintOnScreen flag/attribute. When it's on, no PaintEvent is transmitted to the CellWidget when it's the central widget of the <CellViewer> (QMainWindow). It's something I still don't understand from the doc of Qt. * Change: In <AreaCommand>, use the PlaneId enumeration instead of a anonymous numerical index. * Change: In <HApplication>, no longer catch and silently discard standartd exceptions but launch the ExceptionWidget... Suppress the now deprecated constructor with "Type" argument. * Change: In <SelectionModel>, the "reset()" method is deprecated in Qt5, instead enclose the "clear()" by a "beginResetModel()" and "endResetModel()" pair. * New: In <crlcore>, add support for ISPD05 benchmarks (in Bookshelf format). Forked from ISPD04 and not finished yet. * Change: In <Mauka>, distinguish the Action string identifier from <Etesian> * New: In <unicorn>, add entry for ISPD05 loader. Add entry for <Etesian> analytic placer.
2014-03-22 05:50:36 -05:00
{
size_t w = label.size();
string indent ( w, ' ' );
if (not w) {
label = string( 5, ' ' );
indent = label;
}
Apaired segments building for symmetric routing (step 1). * Change: In Hurricane::BaseFlags, store flags in uint64_t instead of unsigned int because we start to need more than 32 different flags in some tools. * New: In ::getString() & ::getRecord() templates, add support for std::array<>. * Change: In CRL::ToolEngine, add support for timer (time & memory measurements) displaced from Katabatic. This way all ToolEngine can use this feature. The _postCreate() method display the memory just after ToolEngine allocation. * Change: In Etesian::EtesianEngine, make use of the ToolEngine builtin timer (remove the local one). Forgot to call the base class _postCreate() and _preDestroy(). * Change: In Anabatic::AnabaticEngine, make use of the ToolEngine builtin timer (remove the local one). * New: In Anabatic, new AutoSegments_Connecteds() collection. This Collection allows a deterministic walkthough *all* the AutoSegments connected either to source or target of one AutoSegment. * New: In Anabatic::AutoContactTerminal::isEndPoint() to check if an AutoContactTerminal is the *only one* anchored on a RoutingPad, thus being a true "end point" and not a kind of feed-through. * New: In Katana::KatanaEngine, added support for symmetric nets. Created new class DataSymmetric to store symmetric information of a net (mainly the paired AutoSegments). Added KatanaEngine::runSymmetricRouter(), for now only build the DataSymmetric informations. More to come... * Change: In Katana::GraphicKatanaEngine::_runTest(), now perform symmetric information building the non-symmetric routing.
2017-03-12 13:34:12 -05:00
cmess2 << label
<< " HPWL:" << setw(11) << coloquinte::gp::get_HPWL_wirelength( _circuit, _placementLB )
<< " RMST:" << setw(11) << coloquinte::gp::get_RSMT_wirelength( _circuit, _placementLB )
<< endl;
}
void EtesianEngine::_updatePlacement ( const coloquinte::placement_t& placement )
{
Update to Qt 5, requires cmake 2.8.9. New placer: Etesian. Update to Qt 5: * Change: Now requires at least cmake 2.8.9. * Change: CMakeLists.txt needs small changes. Qt modules must be found one by one (Core, Gui, Widgets). Must add "set(CMAKE_AUTOMOC ON)" in the top file and replace "qt4" prefix in macros by "qt5". Added simpler macro "setup_qt()" in FindBootstrap.cmake. * Change: No longer need to include <QGtkStyle> is is choosen by default according to the current desktop environment. * Change: In <hurricane>, In HApplication, launch ExceptionWidget when a std::exception is catched instead of silently discarting it. New placer Tool: Etesian * New: <etesian> analytical placer. Encapsulate Coloquinte from Gabriel Gouvine. * New: in <documentation>, add stub demonstration ToolEngine <smurf>. Needs to be commented. Miscellaneous: * New: in <boostrap> and <unicorn>, added support for Etesian, the new analytic placer. The tool itself will be added in the next commit. * Bug: in <CellWidget>, when shifting the display buffer, we no longer can copy the buffer on itself (we should never have). Now go through a temporary one (PlaneId::AutoCopy) which is added to the DrawingPlanes. Affect "goLeft()" and "goUp()". * Bug: In <CellWidget>, remove the WA_PaintOnScreen flag/attribute. When it's on, no PaintEvent is transmitted to the CellWidget when it's the central widget of the <CellViewer> (QMainWindow). It's something I still don't understand from the doc of Qt. * Change: In <AreaCommand>, use the PlaneId enumeration instead of a anonymous numerical index. * Change: In <HApplication>, no longer catch and silently discard standartd exceptions but launch the ExceptionWidget... Suppress the now deprecated constructor with "Type" argument. * Change: In <SelectionModel>, the "reset()" method is deprecated in Qt5, instead enclose the "clear()" by a "beginResetModel()" and "endResetModel()" pair. * New: In <crlcore>, add support for ISPD05 benchmarks (in Bookshelf format). Forked from ISPD04 and not finished yet. * Change: In <Mauka>, distinguish the Action string identifier from <Etesian> * New: In <unicorn>, add entry for ISPD05 loader. Add entry for <Etesian> analytic placer.
2014-03-22 05:50:36 -05:00
UpdateSession::open();
for ( Occurrence occurrence : getCell()->getLeafInstanceOccurrences() )
Update to Qt 5, requires cmake 2.8.9. New placer: Etesian. Update to Qt 5: * Change: Now requires at least cmake 2.8.9. * Change: CMakeLists.txt needs small changes. Qt modules must be found one by one (Core, Gui, Widgets). Must add "set(CMAKE_AUTOMOC ON)" in the top file and replace "qt4" prefix in macros by "qt5". Added simpler macro "setup_qt()" in FindBootstrap.cmake. * Change: No longer need to include <QGtkStyle> is is choosen by default according to the current desktop environment. * Change: In <hurricane>, In HApplication, launch ExceptionWidget when a std::exception is catched instead of silently discarting it. New placer Tool: Etesian * New: <etesian> analytical placer. Encapsulate Coloquinte from Gabriel Gouvine. * New: in <documentation>, add stub demonstration ToolEngine <smurf>. Needs to be commented. Miscellaneous: * New: in <boostrap> and <unicorn>, added support for Etesian, the new analytic placer. The tool itself will be added in the next commit. * Bug: in <CellWidget>, when shifting the display buffer, we no longer can copy the buffer on itself (we should never have). Now go through a temporary one (PlaneId::AutoCopy) which is added to the DrawingPlanes. Affect "goLeft()" and "goUp()". * Bug: In <CellWidget>, remove the WA_PaintOnScreen flag/attribute. When it's on, no PaintEvent is transmitted to the CellWidget when it's the central widget of the <CellViewer> (QMainWindow). It's something I still don't understand from the doc of Qt. * Change: In <AreaCommand>, use the PlaneId enumeration instead of a anonymous numerical index. * Change: In <HApplication>, no longer catch and silently discard standartd exceptions but launch the ExceptionWidget... Suppress the now deprecated constructor with "Type" argument. * Change: In <SelectionModel>, the "reset()" method is deprecated in Qt5, instead enclose the "clear()" by a "beginResetModel()" and "endResetModel()" pair. * New: In <crlcore>, add support for ISPD05 benchmarks (in Bookshelf format). Forked from ISPD04 and not finished yet. * Change: In <Mauka>, distinguish the Action string identifier from <Etesian> * New: In <unicorn>, add entry for ISPD05 loader. Add entry for <Etesian> analytic placer.
2014-03-22 05:50:36 -05:00
{
DbU::Unit hpitch = getHorizontalPitch();
DbU::Unit vpitch = getVerticalPitch();
Update to Qt 5, requires cmake 2.8.9. New placer: Etesian. Update to Qt 5: * Change: Now requires at least cmake 2.8.9. * Change: CMakeLists.txt needs small changes. Qt modules must be found one by one (Core, Gui, Widgets). Must add "set(CMAKE_AUTOMOC ON)" in the top file and replace "qt4" prefix in macros by "qt5". Added simpler macro "setup_qt()" in FindBootstrap.cmake. * Change: No longer need to include <QGtkStyle> is is choosen by default according to the current desktop environment. * Change: In <hurricane>, In HApplication, launch ExceptionWidget when a std::exception is catched instead of silently discarting it. New placer Tool: Etesian * New: <etesian> analytical placer. Encapsulate Coloquinte from Gabriel Gouvine. * New: in <documentation>, add stub demonstration ToolEngine <smurf>. Needs to be commented. Miscellaneous: * New: in <boostrap> and <unicorn>, added support for Etesian, the new analytic placer. The tool itself will be added in the next commit. * Bug: in <CellWidget>, when shifting the display buffer, we no longer can copy the buffer on itself (we should never have). Now go through a temporary one (PlaneId::AutoCopy) which is added to the DrawingPlanes. Affect "goLeft()" and "goUp()". * Bug: In <CellWidget>, remove the WA_PaintOnScreen flag/attribute. When it's on, no PaintEvent is transmitted to the CellWidget when it's the central widget of the <CellViewer> (QMainWindow). It's something I still don't understand from the doc of Qt. * Change: In <AreaCommand>, use the PlaneId enumeration instead of a anonymous numerical index. * Change: In <HApplication>, no longer catch and silently discard standartd exceptions but launch the ExceptionWidget... Suppress the now deprecated constructor with "Type" argument. * Change: In <SelectionModel>, the "reset()" method is deprecated in Qt5, instead enclose the "clear()" by a "beginResetModel()" and "endResetModel()" pair. * New: In <crlcore>, add support for ISPD05 benchmarks (in Bookshelf format). Forked from ISPD04 and not finished yet. * Change: In <Mauka>, distinguish the Action string identifier from <Etesian> * New: In <unicorn>, add entry for ISPD05 loader. Add entry for <Etesian> analytic placer.
2014-03-22 05:50:36 -05:00
Point instancePosition;
Instance* instance = static_cast<Instance*>(occurrence.getEntity());
string instanceName = occurrence.getCompactString();
Update to Qt 5, requires cmake 2.8.9. New placer: Etesian. Update to Qt 5: * Change: Now requires at least cmake 2.8.9. * Change: CMakeLists.txt needs small changes. Qt modules must be found one by one (Core, Gui, Widgets). Must add "set(CMAKE_AUTOMOC ON)" in the top file and replace "qt4" prefix in macros by "qt5". Added simpler macro "setup_qt()" in FindBootstrap.cmake. * Change: No longer need to include <QGtkStyle> is is choosen by default according to the current desktop environment. * Change: In <hurricane>, In HApplication, launch ExceptionWidget when a std::exception is catched instead of silently discarting it. New placer Tool: Etesian * New: <etesian> analytical placer. Encapsulate Coloquinte from Gabriel Gouvine. * New: in <documentation>, add stub demonstration ToolEngine <smurf>. Needs to be commented. Miscellaneous: * New: in <boostrap> and <unicorn>, added support for Etesian, the new analytic placer. The tool itself will be added in the next commit. * Bug: in <CellWidget>, when shifting the display buffer, we no longer can copy the buffer on itself (we should never have). Now go through a temporary one (PlaneId::AutoCopy) which is added to the DrawingPlanes. Affect "goLeft()" and "goUp()". * Bug: In <CellWidget>, remove the WA_PaintOnScreen flag/attribute. When it's on, no PaintEvent is transmitted to the CellWidget when it's the central widget of the <CellViewer> (QMainWindow). It's something I still don't understand from the doc of Qt. * Change: In <AreaCommand>, use the PlaneId enumeration instead of a anonymous numerical index. * Change: In <HApplication>, no longer catch and silently discard standartd exceptions but launch the ExceptionWidget... Suppress the now deprecated constructor with "Type" argument. * Change: In <SelectionModel>, the "reset()" method is deprecated in Qt5, instead enclose the "clear()" by a "beginResetModel()" and "endResetModel()" pair. * New: In <crlcore>, add support for ISPD05 benchmarks (in Bookshelf format). Forked from ISPD04 and not finished yet. * Change: In <Mauka>, distinguish the Action string identifier from <Etesian> * New: In <unicorn>, add entry for ISPD05 loader. Add entry for <Etesian> analytic placer.
2014-03-22 05:50:36 -05:00
// Remove the enclosing brackets...
instanceName.erase( 0, 1 );
instanceName.erase( instanceName.size()-1 );
auto iid = _cellsToIds.find( instanceName );
if (iid == _cellsToIds.end() ) {
cerr << Error( "Unable to lookup instance <%s>.", instanceName.c_str() ) << endl;
} else {
Etesian integration with Chip & ClockTree plugins. * New: In Hurricane, in DebugSession, add a new method to activate the trace inconditionally with a certain level. * New: In Hurricane, in HyperNet, allow copy construction as there is no reason to disallow it and we need it now. * New: In Hurricane, in Cell::flattenNets(), add a new option to prevent the flattening of the clock net(s). For more safety perform the DeepNet creation *outside* the Collection loop. * Bug: In Hurricane, in Cell_HyperNetRootNetOccurrences, skip the DeepNets because they are the result of another flattening operation. * New: In Isobar, in PyBasicLayer, export C++ method getBlockageLayer(). * New: In Isobar, in PyRoutingGauge, export C++ method getLayerPitch(), needed by Cumulus plugins. * New: In Etesian, EtesianEngine::findYSpin() to look for the Y orientation if some cells are already placed. Typically the buffers of a clock tree. Pass the correct orienation to row_compatible_orientation(). Do not try to add feeds in the ISPD05 benchmarks. For now the benchmarks are detected through their names (unreliable). * Change: In Knik, in KnikEngine::initGlobalRouting(), allow the clock to be routed as an ordinary signal when the clock tree is not used. * New: In Kite, in BuildPowerRails, management & detection for the pre-routed clock. In KiteEngine constructor, early initialization of the blockage net to avoid later troubles in BuildPowerRails. * New: In Cumulus, in ChipPlugin, add support for Etesian plus new configuration parameter 'clockTree.placerEngine' to select between Mauka and Etesian. * New: In Cumulus, in BlockCorona, add blockages in the vertical sides in the vertical layer to prevent the router to use the vertical tracks in under the prower lines (for example, blockage in M3 with power line M5). In Cumulus, in ChipConf add attribute to access the blockage net. * New: In Cumulus, when the clock tree is disabled, do not generate the last rail around the block (the clock rail). * Bug: In Cumulus, in ChipConf use the clock pad to guess the clock signals and *not* the power pad. Add more meaningful error messages if a pad global signal is not found (implicit connexion by name). * Bug: In Cumulus, in ClockTree, compute correctly the cells Y spin, that is *from the bottom of the cell AB* (not from 'zero').
2015-02-24 06:13:17 -06:00
if (instance->getPlacementStatus() == Instance::PlacementStatus::FIXED)
continue;
Update to Qt 5, requires cmake 2.8.9. New placer: Etesian. Update to Qt 5: * Change: Now requires at least cmake 2.8.9. * Change: CMakeLists.txt needs small changes. Qt modules must be found one by one (Core, Gui, Widgets). Must add "set(CMAKE_AUTOMOC ON)" in the top file and replace "qt4" prefix in macros by "qt5". Added simpler macro "setup_qt()" in FindBootstrap.cmake. * Change: No longer need to include <QGtkStyle> is is choosen by default according to the current desktop environment. * Change: In <hurricane>, In HApplication, launch ExceptionWidget when a std::exception is catched instead of silently discarting it. New placer Tool: Etesian * New: <etesian> analytical placer. Encapsulate Coloquinte from Gabriel Gouvine. * New: in <documentation>, add stub demonstration ToolEngine <smurf>. Needs to be commented. Miscellaneous: * New: in <boostrap> and <unicorn>, added support for Etesian, the new analytic placer. The tool itself will be added in the next commit. * Bug: in <CellWidget>, when shifting the display buffer, we no longer can copy the buffer on itself (we should never have). Now go through a temporary one (PlaneId::AutoCopy) which is added to the DrawingPlanes. Affect "goLeft()" and "goUp()". * Bug: In <CellWidget>, remove the WA_PaintOnScreen flag/attribute. When it's on, no PaintEvent is transmitted to the CellWidget when it's the central widget of the <CellViewer> (QMainWindow). It's something I still don't understand from the doc of Qt. * Change: In <AreaCommand>, use the PlaneId enumeration instead of a anonymous numerical index. * Change: In <HApplication>, no longer catch and silently discard standartd exceptions but launch the ExceptionWidget... Suppress the now deprecated constructor with "Type" argument. * Change: In <SelectionModel>, the "reset()" method is deprecated in Qt5, instead enclose the "clear()" by a "beginResetModel()" and "endResetModel()" pair. * New: In <crlcore>, add support for ISPD05 benchmarks (in Bookshelf format). Forked from ISPD04 and not finished yet. * Change: In <Mauka>, distinguish the Action string identifier from <Etesian> * New: In <unicorn>, add entry for ISPD05 loader. Add entry for <Etesian> analytic placer.
2014-03-22 05:50:36 -05:00
Etesian integration with Chip & ClockTree plugins. * New: In Hurricane, in DebugSession, add a new method to activate the trace inconditionally with a certain level. * New: In Hurricane, in HyperNet, allow copy construction as there is no reason to disallow it and we need it now. * New: In Hurricane, in Cell::flattenNets(), add a new option to prevent the flattening of the clock net(s). For more safety perform the DeepNet creation *outside* the Collection loop. * Bug: In Hurricane, in Cell_HyperNetRootNetOccurrences, skip the DeepNets because they are the result of another flattening operation. * New: In Isobar, in PyBasicLayer, export C++ method getBlockageLayer(). * New: In Isobar, in PyRoutingGauge, export C++ method getLayerPitch(), needed by Cumulus plugins. * New: In Etesian, EtesianEngine::findYSpin() to look for the Y orientation if some cells are already placed. Typically the buffers of a clock tree. Pass the correct orienation to row_compatible_orientation(). Do not try to add feeds in the ISPD05 benchmarks. For now the benchmarks are detected through their names (unreliable). * Change: In Knik, in KnikEngine::initGlobalRouting(), allow the clock to be routed as an ordinary signal when the clock tree is not used. * New: In Kite, in BuildPowerRails, management & detection for the pre-routed clock. In KiteEngine constructor, early initialization of the blockage net to avoid later troubles in BuildPowerRails. * New: In Cumulus, in ChipPlugin, add support for Etesian plus new configuration parameter 'clockTree.placerEngine' to select between Mauka and Etesian. * New: In Cumulus, in BlockCorona, add blockages in the vertical sides in the vertical layer to prevent the router to use the vertical tracks in under the prower lines (for example, blockage in M3 with power line M5). In Cumulus, in ChipConf add attribute to access the blockage net. * New: In Cumulus, when the clock tree is disabled, do not generate the last rail around the block (the clock rail). * Bug: In Cumulus, in ChipConf use the clock pad to guess the clock signals and *not* the power pad. Add more meaningful error messages if a pad global signal is not found (implicit connexion by name). * Bug: In Cumulus, in ClockTree, compute correctly the cells Y spin, that is *from the bottom of the cell AB* (not from 'zero').
2015-02-24 06:13:17 -06:00
point<int_t> position = placement.positions_[(*iid).second];
Transformation trans = toTransformation( position
, placement.orientations_[(*iid).second]
, instance->getMasterCell()
, hpitch
, vpitch
Etesian integration with Chip & ClockTree plugins. * New: In Hurricane, in DebugSession, add a new method to activate the trace inconditionally with a certain level. * New: In Hurricane, in HyperNet, allow copy construction as there is no reason to disallow it and we need it now. * New: In Hurricane, in Cell::flattenNets(), add a new option to prevent the flattening of the clock net(s). For more safety perform the DeepNet creation *outside* the Collection loop. * Bug: In Hurricane, in Cell_HyperNetRootNetOccurrences, skip the DeepNets because they are the result of another flattening operation. * New: In Isobar, in PyBasicLayer, export C++ method getBlockageLayer(). * New: In Isobar, in PyRoutingGauge, export C++ method getLayerPitch(), needed by Cumulus plugins. * New: In Etesian, EtesianEngine::findYSpin() to look for the Y orientation if some cells are already placed. Typically the buffers of a clock tree. Pass the correct orienation to row_compatible_orientation(). Do not try to add feeds in the ISPD05 benchmarks. For now the benchmarks are detected through their names (unreliable). * Change: In Knik, in KnikEngine::initGlobalRouting(), allow the clock to be routed as an ordinary signal when the clock tree is not used. * New: In Kite, in BuildPowerRails, management & detection for the pre-routed clock. In KiteEngine constructor, early initialization of the blockage net to avoid later troubles in BuildPowerRails. * New: In Cumulus, in ChipPlugin, add support for Etesian plus new configuration parameter 'clockTree.placerEngine' to select between Mauka and Etesian. * New: In Cumulus, in BlockCorona, add blockages in the vertical sides in the vertical layer to prevent the router to use the vertical tracks in under the prower lines (for example, blockage in M3 with power line M5). In Cumulus, in ChipConf add attribute to access the blockage net. * New: In Cumulus, when the clock tree is disabled, do not generate the last rail around the block (the clock rail). * Bug: In Cumulus, in ChipConf use the clock pad to guess the clock signals and *not* the power pad. Add more meaningful error messages if a pad global signal is not found (implicit connexion by name). * Bug: In Cumulus, in ClockTree, compute correctly the cells Y spin, that is *from the bottom of the cell AB* (not from 'zero').
2015-02-24 06:13:17 -06:00
);
//cerr << "Setting <" << instanceName << " @" << instancePosition << endl;
Update to Qt 5, requires cmake 2.8.9. New placer: Etesian. Update to Qt 5: * Change: Now requires at least cmake 2.8.9. * Change: CMakeLists.txt needs small changes. Qt modules must be found one by one (Core, Gui, Widgets). Must add "set(CMAKE_AUTOMOC ON)" in the top file and replace "qt4" prefix in macros by "qt5". Added simpler macro "setup_qt()" in FindBootstrap.cmake. * Change: No longer need to include <QGtkStyle> is is choosen by default according to the current desktop environment. * Change: In <hurricane>, In HApplication, launch ExceptionWidget when a std::exception is catched instead of silently discarting it. New placer Tool: Etesian * New: <etesian> analytical placer. Encapsulate Coloquinte from Gabriel Gouvine. * New: in <documentation>, add stub demonstration ToolEngine <smurf>. Needs to be commented. Miscellaneous: * New: in <boostrap> and <unicorn>, added support for Etesian, the new analytic placer. The tool itself will be added in the next commit. * Bug: in <CellWidget>, when shifting the display buffer, we no longer can copy the buffer on itself (we should never have). Now go through a temporary one (PlaneId::AutoCopy) which is added to the DrawingPlanes. Affect "goLeft()" and "goUp()". * Bug: In <CellWidget>, remove the WA_PaintOnScreen flag/attribute. When it's on, no PaintEvent is transmitted to the CellWidget when it's the central widget of the <CellViewer> (QMainWindow). It's something I still don't understand from the doc of Qt. * Change: In <AreaCommand>, use the PlaneId enumeration instead of a anonymous numerical index. * Change: In <HApplication>, no longer catch and silently discard standartd exceptions but launch the ExceptionWidget... Suppress the now deprecated constructor with "Type" argument. * Change: In <SelectionModel>, the "reset()" method is deprecated in Qt5, instead enclose the "clear()" by a "beginResetModel()" and "endResetModel()" pair. * New: In <crlcore>, add support for ISPD05 benchmarks (in Bookshelf format). Forked from ISPD04 and not finished yet. * Change: In <Mauka>, distinguish the Action string identifier from <Etesian> * New: In <unicorn>, add entry for ISPD05 loader. Add entry for <Etesian> analytic placer.
2014-03-22 05:50:36 -05:00
Implementation of DataBase native save/restore in JSON (step 1). * New: In Hurricane, added first support for DataBase native import/export using JSON. We choose RapidJSON, in SAX mode, to manage the JSON format low level Read/Write. Thus, it's Git repository http://github.com/miloyip/rapidjson must be added under ~/coriolis-2.x/src and manually build and installed in the Coriolis installation tree (to be integrated in ccb later). Two mode are being supported: 1. Cell mode: one Cell only is saved. In that mode, Entities referred by Occurrences are "outside" the file. They are coded through their "signature" (mostly, all the values of their attributes). The ids saved in the file cannot be restored identically as we cannot predict when and in which context the Cell will be reloaded. 2. Design Blob mode: the whole design hierarchy, down and including the standard cells is saved. This way the design is completly self contained and Entities ared referred through their ids. A design blob can only be loaded immediatly after starting cgt as the DataBase *must* be empty. This way we restore the whole design hierarchy with *exactly* the same ids. Now, Hurricane object should provide a "toJson()" method for driving JSON, and be associated with a JsonObject derived class for parsing. * New: In Hurricane, ability to force the next id that will be used for a DBo (used by Design Blob Mode). * New: In Hurricane, in DataBase, added getCell() and getLibrary() functions to allow the hierarchical access of a Cell/Library in native mode (i.e. whithout the requirement of AllianceFramework). * New: In Hurricane, In CellViewer, added menu entry for Save/Load of JSON Design Blobs. Added at this level because we consider it as the "native" format of Hurricane. * New: In Unicorn, added support of import/export of JSON Cell. * Bug: In Hurricane, in Instance, when cloning an Instance or uniquifying it's master Cell, we forgot about the Occurrences (through shared pathes). When an instance is cloned the Shared pathes still points toward the original Instance. And when it's the master Cell that is uniquifyed it's the Entities pointed to that remains in the original Cell. This is a software design problem. It is difficult to define what policy to adopt when uniquifying: basically that means that one Occurence is either moved onto the clone or duplicated. Furthermore, it is not trivial to known what Occurrence is pointing on the uniquifyed/cloned item. Have to think about it a little more. * Bug: In Etesian, in EtesianEngine, build the flattened nets and their RoutingPads *after* uniquifying (through slaving bounding boxes). This way we avoid the Occurrences problem described above. * Bug: In Etesian, in EtesianEngine, invalidate the RoutingPad after processing the placement so they are put into the right quadtree. This problem is due to the fact that the RoutingPads do not belong to the Instance that they refer. And when this instance is moved around, she doesn't inform the RoutingPad that is has moved. More software architecture design to review...
2016-01-07 06:13:16 -06:00
// This is temporary as it's not trans-hierarchic: we ignore the positions
Update to Qt 5, requires cmake 2.8.9. New placer: Etesian. Update to Qt 5: * Change: Now requires at least cmake 2.8.9. * Change: CMakeLists.txt needs small changes. Qt modules must be found one by one (Core, Gui, Widgets). Must add "set(CMAKE_AUTOMOC ON)" in the top file and replace "qt4" prefix in macros by "qt5". Added simpler macro "setup_qt()" in FindBootstrap.cmake. * Change: No longer need to include <QGtkStyle> is is choosen by default according to the current desktop environment. * Change: In <hurricane>, In HApplication, launch ExceptionWidget when a std::exception is catched instead of silently discarting it. New placer Tool: Etesian * New: <etesian> analytical placer. Encapsulate Coloquinte from Gabriel Gouvine. * New: in <documentation>, add stub demonstration ToolEngine <smurf>. Needs to be commented. Miscellaneous: * New: in <boostrap> and <unicorn>, added support for Etesian, the new analytic placer. The tool itself will be added in the next commit. * Bug: in <CellWidget>, when shifting the display buffer, we no longer can copy the buffer on itself (we should never have). Now go through a temporary one (PlaneId::AutoCopy) which is added to the DrawingPlanes. Affect "goLeft()" and "goUp()". * Bug: In <CellWidget>, remove the WA_PaintOnScreen flag/attribute. When it's on, no PaintEvent is transmitted to the CellWidget when it's the central widget of the <CellViewer> (QMainWindow). It's something I still don't understand from the doc of Qt. * Change: In <AreaCommand>, use the PlaneId enumeration instead of a anonymous numerical index. * Change: In <HApplication>, no longer catch and silently discard standartd exceptions but launch the ExceptionWidget... Suppress the now deprecated constructor with "Type" argument. * Change: In <SelectionModel>, the "reset()" method is deprecated in Qt5, instead enclose the "clear()" by a "beginResetModel()" and "endResetModel()" pair. * New: In <crlcore>, add support for ISPD05 benchmarks (in Bookshelf format). Forked from ISPD04 and not finished yet. * Change: In <Mauka>, distinguish the Action string identifier from <Etesian> * New: In <unicorn>, add entry for ISPD05 loader. Add entry for <Etesian> analytic placer.
2014-03-22 05:50:36 -05:00
// of all the intermediary instances.
instance->setTransformation( trans );
instance->setPlacementStatus( Instance::PlacementStatus::PLACED );
Update to Qt 5, requires cmake 2.8.9. New placer: Etesian. Update to Qt 5: * Change: Now requires at least cmake 2.8.9. * Change: CMakeLists.txt needs small changes. Qt modules must be found one by one (Core, Gui, Widgets). Must add "set(CMAKE_AUTOMOC ON)" in the top file and replace "qt4" prefix in macros by "qt5". Added simpler macro "setup_qt()" in FindBootstrap.cmake. * Change: No longer need to include <QGtkStyle> is is choosen by default according to the current desktop environment. * Change: In <hurricane>, In HApplication, launch ExceptionWidget when a std::exception is catched instead of silently discarting it. New placer Tool: Etesian * New: <etesian> analytical placer. Encapsulate Coloquinte from Gabriel Gouvine. * New: in <documentation>, add stub demonstration ToolEngine <smurf>. Needs to be commented. Miscellaneous: * New: in <boostrap> and <unicorn>, added support for Etesian, the new analytic placer. The tool itself will be added in the next commit. * Bug: in <CellWidget>, when shifting the display buffer, we no longer can copy the buffer on itself (we should never have). Now go through a temporary one (PlaneId::AutoCopy) which is added to the DrawingPlanes. Affect "goLeft()" and "goUp()". * Bug: In <CellWidget>, remove the WA_PaintOnScreen flag/attribute. When it's on, no PaintEvent is transmitted to the CellWidget when it's the central widget of the <CellViewer> (QMainWindow). It's something I still don't understand from the doc of Qt. * Change: In <AreaCommand>, use the PlaneId enumeration instead of a anonymous numerical index. * Change: In <HApplication>, no longer catch and silently discard standartd exceptions but launch the ExceptionWidget... Suppress the now deprecated constructor with "Type" argument. * Change: In <SelectionModel>, the "reset()" method is deprecated in Qt5, instead enclose the "clear()" by a "beginResetModel()" and "endResetModel()" pair. * New: In <crlcore>, add support for ISPD05 benchmarks (in Bookshelf format). Forked from ISPD04 and not finished yet. * Change: In <Mauka>, distinguish the Action string identifier from <Etesian> * New: In <unicorn>, add entry for ISPD05 loader. Add entry for <Etesian> analytic placer.
2014-03-22 05:50:36 -05:00
}
}
UpdateSession::close();
Better handling of exceptions between C++ & Python. Misc. checks. * Bug: In Hurricane, in PyHurricane.h the macro HCATCH was not catching standard STL exceptions. This was the source of the cryptic message: "Fatal Python error: Py_EndInterpreter: thread still has a frame" The Python interpreter was interrupted uncleanly bypassing it's own exceptions mechanism. In PyViewer, the Viewer *do not* inherit from a base class (in the Python export). * New: In Hurricane, in DbU, compute maximum values (in double) for grid, lambda & physical (in meter) so now the DbU::toGrid(), DbU::toLambda() & DbU::toPhysical() methods can check for out of bound values, and throw an exception. * Change: In Hurricane, ExceptionWidget::catchAllWrapper() now returns a boolean, set to <true> if an exception has been catched. Allow callers to interrupt themselves if a problem has occured. * Bug: In Kite & Etesian, in the Python wrapper, send a Python exception if catchAllwrapper() did return true, instead of continuing... * Change: In Kite & Etesian, adds a setViewer() method (exported in Python) to use the graphical ExceptionWidget when in graphic mode. * Bug: In Cumulus, in PadsCorona.py the check for the core vs. chip size was not returning False when invalid. * New: In CRL Core, in Vst driver, add a support IEEE VHDL. Inactive for now as I don't see clearly the policy for selecting it or not. Remove the code of the old Vst driver. In Blif parser, check for non-existent models (incomplete or corrupted Blif file). Found by G. Gouvine. * New: Added extras file for IDE-like support under Emacs.
2015-05-20 07:02:18 -05:00
if (_viewer) _viewer->getCellWidget()->refresh();
Update to Qt 5, requires cmake 2.8.9. New placer: Etesian. Update to Qt 5: * Change: Now requires at least cmake 2.8.9. * Change: CMakeLists.txt needs small changes. Qt modules must be found one by one (Core, Gui, Widgets). Must add "set(CMAKE_AUTOMOC ON)" in the top file and replace "qt4" prefix in macros by "qt5". Added simpler macro "setup_qt()" in FindBootstrap.cmake. * Change: No longer need to include <QGtkStyle> is is choosen by default according to the current desktop environment. * Change: In <hurricane>, In HApplication, launch ExceptionWidget when a std::exception is catched instead of silently discarting it. New placer Tool: Etesian * New: <etesian> analytical placer. Encapsulate Coloquinte from Gabriel Gouvine. * New: in <documentation>, add stub demonstration ToolEngine <smurf>. Needs to be commented. Miscellaneous: * New: in <boostrap> and <unicorn>, added support for Etesian, the new analytic placer. The tool itself will be added in the next commit. * Bug: in <CellWidget>, when shifting the display buffer, we no longer can copy the buffer on itself (we should never have). Now go through a temporary one (PlaneId::AutoCopy) which is added to the DrawingPlanes. Affect "goLeft()" and "goUp()". * Bug: In <CellWidget>, remove the WA_PaintOnScreen flag/attribute. When it's on, no PaintEvent is transmitted to the CellWidget when it's the central widget of the <CellViewer> (QMainWindow). It's something I still don't understand from the doc of Qt. * Change: In <AreaCommand>, use the PlaneId enumeration instead of a anonymous numerical index. * Change: In <HApplication>, no longer catch and silently discard standartd exceptions but launch the ExceptionWidget... Suppress the now deprecated constructor with "Type" argument. * Change: In <SelectionModel>, the "reset()" method is deprecated in Qt5, instead enclose the "clear()" by a "beginResetModel()" and "endResetModel()" pair. * New: In <crlcore>, add support for ISPD05 benchmarks (in Bookshelf format). Forked from ISPD04 and not finished yet. * Change: In <Mauka>, distinguish the Action string identifier from <Etesian> * New: In <unicorn>, add entry for ISPD05 loader. Add entry for <Etesian> analytic placer.
2014-03-22 05:50:36 -05:00
}
string EtesianEngine::_getTypeName () const
{ return "Etesian::EtesianEngine"; }
string EtesianEngine::_getString () const
{
ostringstream os;
os << "<" << "EtesianEngine " << getCell()->getName () << ">";
return os.str();
}
Record* EtesianEngine::_getRecord () const
{
Apaired segments building for symmetric routing (step 1). * Change: In Hurricane::BaseFlags, store flags in uint64_t instead of unsigned int because we start to need more than 32 different flags in some tools. * New: In ::getString() & ::getRecord() templates, add support for std::array<>. * Change: In CRL::ToolEngine, add support for timer (time & memory measurements) displaced from Katabatic. This way all ToolEngine can use this feature. The _postCreate() method display the memory just after ToolEngine allocation. * Change: In Etesian::EtesianEngine, make use of the ToolEngine builtin timer (remove the local one). Forgot to call the base class _postCreate() and _preDestroy(). * Change: In Anabatic::AnabaticEngine, make use of the ToolEngine builtin timer (remove the local one). * New: In Anabatic, new AutoSegments_Connecteds() collection. This Collection allows a deterministic walkthough *all* the AutoSegments connected either to source or target of one AutoSegment. * New: In Anabatic::AutoContactTerminal::isEndPoint() to check if an AutoContactTerminal is the *only one* anchored on a RoutingPad, thus being a true "end point" and not a kind of feed-through. * New: In Katana::KatanaEngine, added support for symmetric nets. Created new class DataSymmetric to store symmetric information of a net (mainly the paired AutoSegments). Added KatanaEngine::runSymmetricRouter(), for now only build the DataSymmetric informations. More to come... * Change: In Katana::GraphicKatanaEngine::_runTest(), now perform symmetric information building the non-symmetric routing.
2017-03-12 13:34:12 -05:00
Record* record = Super::_getRecord ();
Update to Qt 5, requires cmake 2.8.9. New placer: Etesian. Update to Qt 5: * Change: Now requires at least cmake 2.8.9. * Change: CMakeLists.txt needs small changes. Qt modules must be found one by one (Core, Gui, Widgets). Must add "set(CMAKE_AUTOMOC ON)" in the top file and replace "qt4" prefix in macros by "qt5". Added simpler macro "setup_qt()" in FindBootstrap.cmake. * Change: No longer need to include <QGtkStyle> is is choosen by default according to the current desktop environment. * Change: In <hurricane>, In HApplication, launch ExceptionWidget when a std::exception is catched instead of silently discarting it. New placer Tool: Etesian * New: <etesian> analytical placer. Encapsulate Coloquinte from Gabriel Gouvine. * New: in <documentation>, add stub demonstration ToolEngine <smurf>. Needs to be commented. Miscellaneous: * New: in <boostrap> and <unicorn>, added support for Etesian, the new analytic placer. The tool itself will be added in the next commit. * Bug: in <CellWidget>, when shifting the display buffer, we no longer can copy the buffer on itself (we should never have). Now go through a temporary one (PlaneId::AutoCopy) which is added to the DrawingPlanes. Affect "goLeft()" and "goUp()". * Bug: In <CellWidget>, remove the WA_PaintOnScreen flag/attribute. When it's on, no PaintEvent is transmitted to the CellWidget when it's the central widget of the <CellViewer> (QMainWindow). It's something I still don't understand from the doc of Qt. * Change: In <AreaCommand>, use the PlaneId enumeration instead of a anonymous numerical index. * Change: In <HApplication>, no longer catch and silently discard standartd exceptions but launch the ExceptionWidget... Suppress the now deprecated constructor with "Type" argument. * Change: In <SelectionModel>, the "reset()" method is deprecated in Qt5, instead enclose the "clear()" by a "beginResetModel()" and "endResetModel()" pair. * New: In <crlcore>, add support for ISPD05 benchmarks (in Bookshelf format). Forked from ISPD04 and not finished yet. * Change: In <Mauka>, distinguish the Action string identifier from <Etesian> * New: In <unicorn>, add entry for ISPD05 loader. Add entry for <Etesian> analytic placer.
2014-03-22 05:50:36 -05:00
if (record) {
record->add( getSlot( "_configuration", _configuration ) );
}
return record;
}
} // Etesian namespace.