memory leaks when instantiating an engine

Hi, if i read correctly you are adding one more engine to O.engines every
100 iterations. Memory has to grow then, this in not a memory leak. The
question is more: does it make sense to accumulate engines? The answer is:
no. :)

Le dim. 15 juil. 2018 14:47, Zhang Wenyue <
<email address hidden>> a écrit :

> New question #670889 on Yade:
> https://answers.launchpad.net/yade/+question/670889
>
> Hi,
>
> I'm also doing some work related to the removal of particles, and I have a
> problem similar to Question #452277.
> https://answers.launchpad.net/yade/+question/452277
>
> I tried to set the flowEngine.dead=1 and then set it to 0 after few
> steps, which didn't work. I tried to slice the list of engines to
> instantiate the flowEngine, which led to memory leak. Do anyone have a way
> to release the memory effectively?
>
> Can anyone give me some suggestion? Thanks a lot.
> I'm running Yade 2018.02b, Ubuntu 18.04.
>
> My code is shown as follow:
> =============================================
>
> from yade import pack,qt,plot
>
> num_spheres=1000# number of spheres
> young=1e6
> compFricDegree = 3 # initial contact friction during the confining phase
> finalFricDegree = 30 # contact friction during the deviatoric loading
> mn,mx=Vector3(0,0,0),Vector3(2,2,6) # corners of the initial packing
> flag=1
> flag2=0
>
>
> O.materials.append(FrictMat(young=young,poisson=0.5,frictionAngle=radians(compFricDegree),density=2600,label='spheres'))
>
> O.materials.append(FrictMat(young=young,poisson=0.5,frictionAngle=0,density=0,label='walls'))
> walls=aabbWalls([mn,mx],thickness=0,material='walls')
> wallIds=O.bodies.append(walls)
>
> sp=pack.SpherePack()
> sp.makeCloud(mn,mx,rMean=0.3,rRelFuzz=0,num=50,periodic=False)
> sp.makeCloud(mn,mx,rMean=0.03,rRelFuzz=0,num=1000,periodic=False)
> sp.toSimulation(material='spheres')
>
> triax=TriaxialStressController(
> thickness = 0,
> stressMask = 7,
> max_vel = 0.005,
> internalCompaction=0, # If true the confining pressure is
> generated by growing particles
> # wall_bottom_activated=False
> )
>
> newton=NewtonIntegrator(damping=0.2)
>
> O.engines=[
> ForceResetter(),
> InsertionSortCollider([Bo1_Sphere_Aabb(),Bo1_Box_Aabb()]),
> InteractionLoop(
> [Ig2_Sphere_Sphere_ScGeom(),Ig2_Box_Sphere_ScGeom()],
> [Ip2_FrictMat_FrictMat_FrictPhys()],
> [Law2_ScGeom_FrictPhys_CundallStrack()],label="iloop"
> ),
> FlowEngine(dead=1,label="flow"),#introduced as a dead engine for
> the moment, see 2nd section
>
> GlobalStiffnessTimeStepper(active=1,timeStepUpdateInterval=100,timestepSafetyCoefficient=0.8),
> triax,
> newton,
> ]
>
> triax.goal1=triax.goal2=triax.goal3=-10000
>
> while 1:
> O.run(1000,1)
> unb=unbalancedForce()
> if unb<0.001 and abs(-10000-triax.meanStress)/10000<0.001:
> break
>
> setContactFriction(radians(finalFricDegree))
>
> ## ______________ Oedometer section _________________
>
> triax.stressMask=7
> triax.goal1=triax.goal2=triax.goal3=-10000
>
> #B. Activate flow engine and set boundary conditions in order to get
> permeability
>
> flow.dead=0
> flow.defTolerance=0.3
> flow.meshUpdateInterval=100
> flow.useSolver=3
> flow.permeabilityFactor=1
> flow.viscosity=10
> flow.bndCondIsPressure=[0,0,0,0,1,1]
> flow.bndCondValue=[0,0,0,0,100,0]
> flow.boundaryUseMaxMin=[0,0,0,0,0,0]
> O.dt=0.1e-4
> O.dynDt=False
>
> newton.damping=0
> triax.wall_back_activated=0
> O.engines=O.engines+[PyRunner(command='addPlotData()',iterPeriod=100),]
> O.engines=O.engines+[PyRunner(command='erosion()',iterPeriod=100),]
>
> def erosion():
> global flag
> global flag2
> k=0
> if flag:
> for i in O.interactions:
> if i.id1==5:
> if i.id2>5 and O.bodies[i.id2].shape.radius<0.1:
> k=i.id2
> O.interactions.erase(i.id1, i.id2)
> O.bodies.erase(k)
> flag2=1
> if i.id2==5:
> if i.id1>5 and O.bodies[i.id1].shape.radius<0.1:
> k=i.id1
> O.interactions.erase(i.id1, i.id2)
> O.bodies.erase(k)
> flag2=1
> if flag2:
> flag=0
> O.pause()
> O.engines=O.engines[0:3]+O.engines[4:]
> O.step()
> O.step()
>
> O.engines=O.engines[0:3]+[FlowEngine(dead=0,label="flow"),]+O.engines[3:]
> flow.defTolerance = 0.3
> flow.meshUpdateInterval = 100
> flow.useSolver = 3
> flow.permeabilityFactor = 1
> flow.viscosity = 10
> flow.bndCondIsPressure = [0, 0, 0, 0, 1, 1]
> flow.bndCondValue = [0, 0, 0, 0, 100, 0]
> flow.boundaryUseMaxMin = [0, 0, 0, 0, 0, 0]
> O.dt = 0.1e-4
> O.dynDt = False
> O.step()
> O.step()
> O.step()
> flag=1
> O.run()
> =============================================
>
> Best,
> Zhang Wenyue
>
> --
> You received this question notification because your team yade-users is
> an answer contact for Yade.
>
> _______________________________________________
> Mailing list: https://launchpad.net/~yade-users
> Post to : <email address hidden>
> Unsubscribe : https://launchpad.net/~yade-users
> More help : https://help.launchpad.net/ListHelp
>
>

Dear, Chareyre

Yes, so the problem is how to delete the old one and free the memory. By Antoine in Question #452277, "del flow" is not effective. I need to instantiate the engine frequently so I can't follow the method of saving and reloading.

Best,
Wenyue

Hi, following #452277 the solution is to implement a "reset" function in the engine itself to handle changes in the list of bodies. It can be implemented on the c++ side if you have the patience to do it.
Bruno