nrnhines

Does the simulation work with --multisend but without --gpu?

> if the "--multisend" flag is appended, the program fails quickly Just a datapoint as my desktop takes 20s to simulate 20ms on 8 cores when configured with -DCMAKE_BUILD_TYPE=Debug. Without...

Perhaps this could be improved. I don't know. The multisend method is an interprocessor spike exchange method, which, under some circumstances, has better performance than MPI_Allgather (followed by, if there...

> Do you mean Yes. I would only disambiguate, or stress a bit more, that on dequeuing on the cpu from the destination cpu priority queue that the cpu copies...

> the cpu copies the spike to a modfile specific buffer on the cpu. That buffer is copied every time step to the gpu. The gpu delivers the contents of...

> ...binqueue... if the query method is moved to GPU, the performance should increase greatly. I agree it could be on the GPU. Whether you can get better performance (at...

2. Note that ```deliver_net_events(nth)``` is called on entry to ```nrn_fixed_step_thread(NrnThread* nth)``` to check thresholds and deliver all (including binqueue) events up to tentry+dt/2. Whereas nrn_deliver_events(nth) is called on exit to...

Yes, "not ideal" in the unlikely event the user experiences confusion if noticeing slightly different results depending on ```nrnivmodl``` and ```nrnivmodl -coreneuron``` and is not aware of the reason. However,...

I'm experimenting with the autotools version. I wonder if this in an ambiguity of symbolic link dates so that different machines or different make versions have different behavior. I.e. my...

Looking at bin/nrnmech_makefile it appears that ``` # Build c files with nocmodl $(MODC_DIR)/%.c: $(MODS_PATH)/%.mod | $(MODC_DIR) ``` can never be skipped because the current folder is x86_64 and MODC_DIR...