Pop 3 Maker Segfault
Nathan Lai
0x00005555557bd6eb in pop3_maker (nx=14, ny=28, nz=<optimized out>, d=..., dm=..., h2d=..., h2diss=..., kph=..., temp=..., u=..., v=..., w=..., cooltime=..., dt=0.013145641993152993, r=..., metal=...,
dx=0.0062499999999999917, t=2172.3355615849296, z=0, procnum=0, d1=1.6730000000000106e-20, x1=3.0857e+18, v1=9778179.1678549927, t1=315570000000, nmax=2945, xstart=0.44375000000000003,
ystart=0.081250000000000003, zstart=0.14375000000000002, ibuff=3, imetal=0, imethod=0, h2crit=0.00050000000000000001, metalcrit=0.0001, odthresh=<optimized out>, starmass=100, level=3, np=0, xp=...,
yp=..., zp=..., up=..., vp=..., wp=..., mp=..., tdp=..., tcp=..., metalf=..., type=..., ctype=5, justburn=0, iradtrans=0) at pop3_maker.F:195
195 if (kph(i,j,k) .gt. tiny) goto 10
(gdb) bt
#0 0x00005555557bd6eb in pop3_maker (nx=14, ny=28, nz=<optimized out>, d=..., dm=..., h2d=..., h2diss=..., kph=..., temp=..., u=..., v=..., w=..., cooltime=..., dt=0.013145641993152993, r=..., metal=...,
dx=0.0062499999999999917, t=2172.3355615849296, z=0, procnum=0, d1=1.6730000000000106e-20, x1=3.0857e+18, v1=9778179.1678549927, t1=315570000000, nmax=2945, xstart=0.44375000000000003,
ystart=0.081250000000000003, zstart=0.14375000000000002, ibuff=3, imetal=0, imethod=0, h2crit=0.00050000000000000001, metalcrit=0.0001, odthresh=<optimized out>, starmass=100, level=3, np=0, xp=...,
yp=..., zp=..., up=..., vp=..., wp=..., mp=..., tdp=..., tcp=..., metalf=..., type=..., ctype=5, justburn=0, iradtrans=0) at pop3_maker.F:195
#1 0x0000555555775089 in grid::StarParticleHandler (this=0x55555beaaff0, SubgridPointer=<optimized out>, level=<optimized out>, level@entry=3, dtLevelAbove=dtLevelAbove@entry=0.035201639594406316,
TopGridTimeStep=0) at Grid_StarParticleHandler.C:961
#2 0x0000555555634c04 in EvolveLevel (MetaData=MetaData@entry=0x7fffffffccb0, LevelArray=LevelArray@entry=0x7fffffffcb20, level=level@entry=3, dtLevelAbove=<optimized out>,
Exterior=Exterior@entry=0x7fffffffd020, ImplicitSolver=ImplicitSolver@entry=0x7fffffffcbf8, dt0=dt0@entry=0.12755533538867594, SiblingGridListStorage=SiblingGridListStorage@entry=0x7fffffffc780)
at EvolveLevel.C:694
#3 0x0000555555635128 in EvolveLevel (MetaData=MetaData@entry=0x7fffffffccb0, LevelArray=LevelArray@entry=0x7fffffffcb20, level=level@entry=2, dtLevelAbove=<optimized out>,
Exterior=Exterior@entry=0x7fffffffd020, ImplicitSolver=ImplicitSolver@entry=0x7fffffffcbf8, dt0=dt0@entry=0.12755533538867594, SiblingGridListStorage=SiblingGridListStorage@entry=0x7fffffffc780)
at EvolveLevel.C:818
#4 0x0000555555635128 in EvolveLevel (MetaData=MetaData@entry=0x7fffffffccb0, LevelArray=LevelArray@entry=0x7fffffffcb20, level=level@entry=1, dtLevelAbove=<optimized out>,
Exterior=Exterior@entry=0x7fffffffd020, ImplicitSolver=ImplicitSolver@entry=0x7fffffffcbf8, dt0=dt0@entry=0.12755533538867594, SiblingGridListStorage=SiblingGridListStorage@entry=0x7fffffffc780)
at EvolveLevel.C:818
#5 0x0000555555635128 in EvolveLevel (MetaData=MetaData@entry=0x7fffffffccb0, LevelArray=LevelArray@entry=0x7fffffffcb20, level=level@entry=0, dtLevelAbove=dtLevelAbove@entry=0.12755533538867594,
Exterior=Exterior@entry=0x7fffffffd020, ImplicitSolver=ImplicitSolver@entry=0x7fffffffcbf8, dt0=dt0@entry=0.12755533538867594, SiblingGridListStorage=SiblingGridListStorage@entry=0x7fffffffc780)
at EvolveLevel.C:818
#6 0x0000555555629fca in EvolveHierarchy (TopGrid=..., MetaData=..., Exterior=Exterior@entry=0x7fffffffd020, ImplicitSolver=0x7fffffffcbf8, LevelArray=LevelArray@entry=0x7fffffffcb20,
Initialdt=<optimized out>) at EvolveHierarchy.C:512
#7 0x000055555559f466 in main (argc=<optimized out>, argv=<optimized out>) at enzo.C:793
Was wondering if anyone had any suggestions. I may be just running out of memory but I hope that is not the issue.
Thanks!
Nathan Lai
John Wise
Hi Nathan,
Sorry to hear that you're having issues with the Pop III star formation routine. It's failing when it's trying to access the LW radiation field. It should only check this field when you have both radiation transport on and MultiSpecies > 1 (H2 cooling). Do you have both of these parameters set like this? Can you send your parameter file?
Thanks,
John
--
You received this message because you are subscribed to the Google Groups "enzo-dev" group.
To unsubscribe from this group and stop receiving emails from it, send an email to enzo-dev+u...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/enzo-dev/3f057ed3-263e-48bb-9fa1-c2c31bd6971fn%40googlegroups.com.
-- John Wise Director, Center for Relativistic Astrophysics Professor, School of Physics Georgia Institute of Technology http://cosmo.gatech.edu
Matthew Abruzzo
--
Nathan Lai
Hi all,
#
# define problem
#
ProblemType = 27 // Collapse test
TopGridRank = 3
TopGridDimensions = 20 20 20
SelfGravity = 1 // gravity on
TopGridGravityBoundary = 0 // periodic
LeftFaceBoundaryCondition = 3 3 3 // periodic
RightFaceBoundaryCondition = 3 3 3
#
# problem parameters
#
CollapseTestRefineAtStart = 1
CollapseTestNumberOfSpheres = 1
CollapseTestUseParticles = 1
#CollapseTestInitialTemperature = // temperature of the background gas
CollapseTestSpherePosition[0] = 0.5 0.5 0.5
CollapseTestSphereVelocity[0] = 0.0 0.0 0.0
CollapseTestSphereRadius[0] = 0.1
#CollapseTestSphereCoreRadius[0] = // only used with sphere type 5
CollapseTestSphereDensity[0] = 1000 // sphere density, the background density is 1
CollapseTestSphereTemperature[0] = 5 // put sphere in pressure equilibrium (rho * T is constant)
CollapseTestSphereType[0] = 1 // constant density
// 1: uniform
// 2: r^-2 power-law
// 3: NFW
// 4: Gaussian
// 5: r^-2 power-law with a core
#
# no cosmology for this run
#
ComovingCoordinates = 0 // Expansion OFF
#
# units
#
DensityUnits = 1.673e-20 // 10^4 g cm^-3
LengthUnits = 3.0857e+18 // 1 pc in cm
TimeUnits = 3.1557e+11 // 10^4 yrs
GravitationalConstant = 1.39698e-3 // 4*pi*G_{cgs}*DensityUnits*TimeUnits^2
#
# set I/O and stop/start parameters
#
StopTime = 10000
dtDataDump = 100
StopCycle = 999999999
NewMovieParticleOn = 1
DataDumpDir = DD
DataDumpName = DD
OutputTemperature = 1 // Output temperature field.
OutputGriddedStarParticle = 1
#
# set hydro parameters
#
Gamma = 1.6667
PPMDiffusionParameter = 0 // diffusion off
DualEnergyFormalism = 1 // use total & internal energy
InterpolationMethod = 1 // SecondOrderA
CourantSafetyNumber = 0.5
FluxCorrection = 1
ConservativeInterpolation = 1
HydroMethod = 0 // PPM
RiemannSolver = 5
ParticleCourantSafetyNumber = 0.8
RiemannSolverFallback = 1
UseMinimumPressureSupport = 0
#
# chemistry/cooling
#
MultiSpecies = 2
RadiativeCooling = 1
RadiativeTransfer = 1
#
# set grid refinement parameters
#
StaticHierarchy = 0
MaximumRefinementLevel = 3
MaximumGravityRefinementLevel = 3
MaximumParticleRefinementLevel = 4
RefineBy = 2
CellFlaggingMethod = 2 4
MinimumOverDensityForRefinement = 8.0 8.0
MinimumMassForRefinementLevelExponent = 0.0 0.0
MinimumEfficiency = 0.35
#
# star formation parameters
#
StarParticleCreation = 8
#StarParticleFeedback = 8
StarMakerOverDensityThreshold = 100
StarMakerMassEfficiency = 0.25
StarMassEjectionFraction = 0.25
StarMetalYield = 0.1
StarEnergyToThermalFeedback = 1e-5
#StarFeedbackDistRadius = 1
#StarFeedbackDistCellStep = 3
John Wise
Hi Nathan,
I made a few changes to your parameter file, and when I run it, one Pop III star particle forms at t ~ 1.71. The main problem was that you set the star particle creation density to the initial density of the cloud. In that case, all of the cells inside the cloud will trigger star particle creation, but there is a hard-coded safety check in the Pop III formation routine to prevent many star particles from forming within some radius.
I changed the SF density criterion from 1000 to 1e5, allowing the cloud to collapse before it forms stars, which is a more realistic setup. I also changed the Courant factor from 0.5 to 0.25, which is needed with feedback, and the StopTime to 2.5 and output cadence to 0.25. I'm attaching the modified parameter file, so you can test it out. The Pop III star particles will have its "particle_type" field equal to 5. When it forms, the stdout should give you the following information
"""
StarParticle[birth]: L3, r = 0.00625 pc, M = 1017, Z2/Z3 = 0/0
mass = 100 (100% cold) SolarMass,
rho = 9.77818e+06 g/cm3, tdyn = 0 Myr
vel = -0.000000 -0.000000 -0.000000 (-0.000000 -0.000000
-0.000000)
pos = 0.500000 0.500000 0.500000
"""
I hope this works for you, and let us know if you have any more questions.
Thanks,
John
To view this discussion on the web visit https://groups.google.com/d/msgid/enzo-dev/0a90aa86-6191-4d0d-a1f9-a4b322ea3c74n%40googlegroups.com.
Nathan Lai
John Wise
Yes, you can use the CollapseTest to simulate a metal-free molecular cloud instead of a cosmological simulation. I'd suggest using a published setup and trying to reproduce their results. Most of the parameters are there to control the initial conditions of the cloud. The only thing that is lacking is a good turbulence generator. The one in CollapseTest only does local random motions, but the MHD CollapseTest generates turbulent conditions from some power spectrum. I'm not too familiar with the latter.
Thanks,
John
To view this discussion on the web visit https://groups.google.com/d/msgid/enzo-dev/db6f2648-5d55-4567-97c1-8320ba0392b5n%40googlegroups.com.
Nathan Lai
John Wise
No worries. When Pop III star particles are created, it merges
any particles together within some radius in a given timestep.
Since the bookkeeping of particle IDs in Enzo makes it difficult
to delete particles, we just set the particle masses to a very
small number. The NumberOfStarParticles parameter is used to set
the star particle IDs, and we only increment it. This is why that
.hierarchy entry is larger than the actual star particles. You
should use the data provided with yt / .cpuXXXX files.
Thanks,
John
To view this discussion on the web visit https://groups.google.com/d/msgid/enzo-dev/d6aefb8f-5773-4e7c-a924-d6cb7b38a3e9n%40googlegroups.com.