CoreNeuron execution procedure as of 13th of August

[ikitayama]

As the option -DADDITIONAL_MECHPATH no longer supported in the master branch,
we just use nrnivmodl-core mod?

@nrnhines describes details NEURON and CoreNeuron running sequences at:

https://github.com/nrnhines/ringtest; but the line:

./coreneuron_x86/bin/coreneuron_exec -e 100 -d coreneuron_data

should be replaced with:

./x86_64/bin/special-core -e 100 -d coreneuron_data

? Looks like it is working for me on JURECA at JSC, so just wanted to make sure I'm on the right track.

[nrnhines]

You are correct. Use nrnivmodl-core. But the special I believe is ./x86_64/special-core without the bin.
I will look through the instructions for ringtest and update them.

[nrnhines]

@pramodk. Am I right about the special-core`` location? I've been using an installation of CoreNEURON as a submodule of NEURON (-DNRN_ENABLE_CORENEURON=ON```) but I realize perhaps an external separate build of CoreNEURON may do things differently.

[pramodk]

I am still working on updating README, here draft work-in-progress : https://github.com/BlueBrain/CoreNeuron/blob/pramodk/readme-update/README.md

[ikitayama]

@pramodk I get (on JURECA):

$ python test.py
Could not get symbol nrn2core_all_weights_return_ from CoreNEURON
NEURON: dlsym returned NULL
 near line 0
 create soma
            ^
        ParallelContext[0].psolve(10)
Traceback (most recent call last):
  File "test.py", line 48, in <module>
    pc.psolve(h.tstop)
RuntimeError: hoc error

test.py is a Python file that is taken straight from your example.

[ikitayama]

I wasn't (again) updating the submodules under nrn master.

$ python test.py

 Duke, Yale, and the BlueBrain Project -- Copyright 1984-2020
 version id unimplemented

 Additional mechanisms from files
 CALIca.mod CALIkahp.mod CALIkc.mod CALIkdr.mod CALIna.mod Ca_conce.mod astrocyte.mod halfgap.mod tN1fAstro.mod tN1fN2.mod tN2fAstro.mod tN2fN1.mod

 Memory (MBs) :             After mk_mech : Max 27.4102, Min 27.4102, Avg 27.4102
 Memory (MBs) :            After MPI_Init : Max 27.4102, Min 27.4102, Avg 27.4102
 Memory (MBs) :          Before nrn_setup : Max 27.4102, Min 27.4102, Avg 27.4102
 Setup Done   : 0.01 seconds
 Memory (MBs) :          After nrn_setup  : Max 27.6680, Min 27.6680, Avg 27.6680
GENERAL PARAMETERS
--mpi=false
--gpu=false
--dt=0.025
--tstop=10

GPU
--nwarp=0
--cell-permute=1

INPUT PARAMETERS
--voltage=1000
--seed=-1
--datpath=.
--filesdat=files.dat
--pattern=
--report-conf=
--restore=

PARALLEL COMPUTATION PARAMETERS
--threading=false
--skip_mpi_finalize=true

SPIKE EXCHANGE
--ms_phases=2
--ms_subintervals=2
--multisend=false
--spk_compress=0
--binqueue=false

CONFIGURATION
--spikebuf=100000
--prcellgid=-1
--forwardskip=0
--celsius=6.3
--mindelay=10
--report-buffer-size=4

OUTPUT PARAMETERS
--dt_io=0.1
--outpath=.
--checkpoint=

 Start time (t) = 0

 Memory (MBs) :  After mk_spikevec_buffer : Max 27.6680, Min 27.6680, Avg 27.6680
 Memory (MBs) :     After nrn_finitialize : Max 27.6680, Min 27.6680, Avg 27.6680

 psolve |========================================================| t: 10.00  ETA: 0h00m00s

Solver Time : 0.00173807


 Simulation Statistics
 Number of cells: 1
 Number of compartments: 5
 Number of presyns: 1
 Number of input presyns: 0
 Number of synapses: 1
 Number of point processes: 2
 Number of transfer (gap) targets: 0
 Number of spikes: 2
 Number of spikes with non negative gid-s: 2