forked from Geant4/geant4
-
Notifications
You must be signed in to change notification settings - Fork 2
/
Copy pathPatch4.10.0-3.txt
174 lines (152 loc) · 7.21 KB
/
Patch4.10.0-3.txt
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
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.