Gabriele Cosmo
175 lines
7.2 KiB
Plaintext
175 lines
7.2 KiB
Plaintext
|
|
Geant4 10.0 - patch-03 Release Notes
|
|
------------------------------------
|
|
|
|
31 October 2014
|
|
|
|
List of fixes included in this public patch since the public release 10.0.p02:
|
|
|
|
o Configuration:
|
|
-------------
|
|
+ CMake:
|
|
o Append existing CMake CXX/EXE_LINKER flags to Geant4 flags when
|
|
setting overall CMake flags. This allows users to override the base
|
|
set if needed, at their own risk.
|
|
Addressing problem report #1663.
|
|
|
|
o Analysis:
|
|
--------
|
|
+ Fixed applying unit to min, max values in H1, H2 managers.
|
|
+ Fixed createH2/setH2 commands implementation.
|
|
|
|
o Geometry:
|
|
--------
|
|
+ management:
|
|
o Fixed possible data-race condition among threads in G4GeomSplitter.
|
|
+ navigation:
|
|
o Fix in G4MultiNavigator to properly set 'fIdNavLimiting' index to
|
|
navigator limiting the step, if only one navigator limits the step.
|
|
o Removed unnecessary thread_local data specifications in
|
|
G4ReplicaNavigation.
|
|
+ solids/Boolean:
|
|
o Fixed use of polyhedron pointer, to avoid race condition in MT
|
|
when parameterised volumes are visualised.
|
|
+ solids/CSG:
|
|
o G4Cons: fixed typo error in GetPointOnSurface() in the call
|
|
to GetRadiusInRing(). Responsible for occasionally wrong reports
|
|
when checking for overlaps.
|
|
o G4Cons: fix in DistanceToIn(p,v) for points on the Surface, to resolve
|
|
error reported in unit testing by the OpticalEscape test.
|
|
o Fixed use of polyhedron pointer, to avoid race condition in MT
|
|
when parameterised volumes are visualised.
|
|
+ solids/specific:
|
|
o Fixed use of polyhedron pointer, to avoid race condition in MT
|
|
when parameterised volumes are visualised.
|
|
+ solids/usolids:
|
|
o UCons: fixed typo error in GetPointOnSurface() in the call
|
|
to GetRadiusInRing(). Responsible for occasionally wrong reports
|
|
when checking for overlaps.
|
|
Fix in DistanceToIn(p,v) for points on the Surface, to resolve
|
|
error reported in unit testing by the OpticalEscape test.
|
|
o USphere: fixed typo in SafetyFromInside().
|
|
o Minor fix in UVector2 for variable shadowing.
|
|
+ volumes:
|
|
o Corrected possible MT data race in G4PVReplica::InitialiseWorker().
|
|
|
|
o Global:
|
|
------
|
|
+ Added symbol to exclusion list in windefs.h.
|
|
+ Use G4Exp, G4Log and G4Pow to speedup G4JTPolynomialSolver.
|
|
+ Updated date and version for 10.0.p03.
|
|
|
|
o Graphics Representations:
|
|
------------------------
|
|
+ HepPolyhedron::SetReferences(): added protection against a null pointer.
|
|
+ BooleanProcessor::execute(): corrected spelling mistake in message.
|
|
|
|
o Particles:
|
|
---------
|
|
+ G4ElectronOccupancy: added G4Exception to protect against unphysical
|
|
orbits.
|
|
+ G4NuclideTable : Fix missing Cl35 ground state in NuclideTable2 and
|
|
missing "*std::log(2.0)" in some comparison of threshold time.
|
|
+ G4IonTable: fill decay channels when an ion is created on the fly to
|
|
avoid race condition issue in multi-threaded mode.
|
|
Addressing problem report #1651.
|
|
+ Fix for possible race condition among threads in G4PDefSplitter.
|
|
|
|
o Physics Lists:
|
|
-------------
|
|
+ constructors/limiters:
|
|
o Fixed possible data-race condition among threads in
|
|
G4NeutronTrackingCut when calling G4MTcout (defect spotted with DRD).
|
|
o G4NeutronTrackingCut: register G4NeutronKiller in
|
|
G4HadronicProcessStore; improved log and process deleted at exit.
|
|
o G4StepLimiterPhysics: fixed minor memory leak reported by Coverity.
|
|
|
|
o Processes - Electromagnetic :
|
|
---------------------------
|
|
+ High Energy:
|
|
o Verified and updated positron annihilation models (G4Vee2hadrons,
|
|
G4ee2KChargedModel, G4ee2KNeutralModel, G4eeTo3PiModel,
|
|
G4eeToHadronsMultiModel, G4eeToPGammaModel, G4eeToTwoPiModel,
|
|
G4eeToHadronsModel) and process G4eeToHadrons.
|
|
This includes addition of extra inline methods to convert from Lab
|
|
to CM, to initialise models with more correct thresholds.
|
|
o G4eeTo3PiModel: fixed sampling algorithm.
|
|
+ Low Energy:
|
|
o Fixed cases of energy violation in G4LivermoreComptonModel and
|
|
atomic de-excitation. Addressing problem report #1612.
|
|
+ Standard:
|
|
o G4PAIxSection: added protection to avoid division by zero.
|
|
Addressing problem report #1677.
|
|
|
|
o Processes - Hadronic:
|
|
--------------------
|
|
+ models/de_excitation:
|
|
o Converted G4FermiFragmentsPool to be a G4ThreadLocalSingleton, to fix
|
|
cases of data race due to lazy initialisation. Effects: sligthly
|
|
increase of memory per thread; no effects in physics.
|
|
+ models/im_r_matrix:
|
|
o Fixed out of bound error in G4DetailedBalancePhaseSpaceIntegral,
|
|
reported by Coverity.
|
|
+ models/inclxx:
|
|
o Fix a bug in Coulomb deflection for negative incident particles.
|
|
o Fix: avoid negative Fermi energies for delta resonances.
|
|
+ models/pre_equilibrium:
|
|
o G4PreCompoundAlpha: fixed integer overflow.
|
|
Addressing problem report #1646.
|
|
+ models/quasi_elastic:
|
|
o Removed potential data-race condition among threads.
|
|
Addressing problem report #1652.
|
|
+ models/radioactive_decay:
|
|
o Fixed memory leak in destructor of G4RIsotopeTable.
|
|
Addressing problem report #1617.
|
|
|
|
o Run:
|
|
-------
|
|
+ Fixed data-race condition among threads when creating RNG for workers.
|
|
+ Fixed '/random/setSeeds' command not working properly.
|
|
Addressing problem report #1657.
|
|
+ Fixed potential race condition among threads in G4VUPLSplitter class.
|
|
+ G4PhysicsListHelper: minor fix in the default list of processes:
|
|
positron annihilation in muons and hadrons cannot be AtRest; these are
|
|
rare high energy discrete processes. Equivalent fix done for
|
|
OrderingParameterTable().
|
|
Addressiong problem report #1598.
|
|
+ Clear root logical volume pointers in G4Region when geometry is
|
|
rebuilt from scratch. Addressing problem report #1665.
|
|
+ G4MTRunManager: make sure G4RunManager::TerminateEventLoop() is
|
|
invoked at the end of run.
|
|
+ Addressing possible data-race in G4RunManagerKernel::SetupPhysics()
|
|
when calling G4ParticleTable::SetReadiness().
|
|
|
|
o Visualization:
|
|
-------------
|
|
+ gMocren:
|
|
o Fixed memory leak in G4GMocrenFile. Addressing problem report #1585.
|
|
|
|
|
|
o Examples:
|
|
--------
|
|
+ Updated reference outputs.
|
|
+ extended/runAndEvent/RE05:
|
|
o Reduced amount of print-out.
|
|
+ extended/runAndEvent/RE06:
|
|
o Added Boolean protection in ConstructSD methods for MT.
|
|
|
|
|
|
----------------------------------------------------------------------------
|
|
|
|
Technical Notes
|
|
---------------
|
|
|
|
o This patch should be applied on top of release 10.0, 10.0.p01 or 10.0.p02.
|
|
o Technical notes distributed for release 10.0 are also applicable and
|
|
valid for this patch.
|
|
|
|
The code and rebuilt binary libraries for release 10.0.p03 are available
|
|
through our "Download" Web page:
|
|
http://cern.ch/geant4/support/download.shtml
|
|
|
|
Please refer to the Geant4 User Documentation:
|
|
http://cern.ch/geant4/support/userdocuments.shtml
|
|
for further information about using Geant4.
|