www.neuron.yale.edu

Hi,

I have tried parallelization of different types of cells by putting them in one for loop and checking with the local variable i whether it has reached the cell count limit. And based upon that it is assigned gids.

Now when I use the below lines:
cvode=new CVode()
cvode.active(1)
cvode.use_daspk(1)
cvode.condition_order(2)

It gives me this error:
2 /home/krishna/nrn-7.3_installed/x86_64/bin/nrniv: spFactor error: Singular
2 in test_mosinit.hoc near line 102
2 cvode.use_daspk(1)
^
2 CVode[1].use_daspk(1)
application called MPI_Abort(MPI_COMM_WORLD, -1) - process 2

However, when I use only one cell type, it doesn't give any problem. The problem seems to arise when two different cell types are created within for loop.

Kindly let me know how I can solve this problem.

Thank you.

What happens if you replace
cvode.active(1)
with
cvode_active(1)
?

Hi,

The problem now seems to be with cvode.use_daspk(1) after I changed cvode.active(1) to cvode_active(1).

1 /home/krishna/nrn-7.3_installed/x86_64/bin/nrniv: spFactor error: Singular
1 in test_mosinit.hoc near line 102
1 cvode.use_daspk(1)
^
1 CVode[1].use_daspk(1)
application called MPI_Abort(MPI_COMM_WORLD, -1) - process 1

How do I resolve this? Kindly let me know.

Thank you.

I don't know which secton of code is to blame given the combination of your (unknown to me) network model, the DAE solver, and condition order 2.
Could you please send me the hoc,mod,ses,py, etc, files in a zip file so I can observe the error. Send to michael dot hines at yale dot edu .
I'm guessing that your model does not require the DAE solver (only needed if extracellular or LinearMechanism is being used) so I'd like to know also if it works without the cvode.use_daspk.