Omar Awile
Omar Awile
While I'm liking opengrok more and more I find the documentation could need some work. There are many small typos and a number of points that could use clarification and...
This relates to #1894 MUSIC support was inadvertently dropped when we switched to CMake. We're fixing this oversight now. This is still a bit work in progress but we have...
As we are moving to a more object-oriented implementation we would like to turn the `Memb_func` struct into a proper C++ class that also encapsulates operations on its data into...
To replace the more C-style `alloc_memory`, `calloc_memory`, `emalloc_align` and `ecalloc_align` with C++ class construction and allocation (using `new` and `delete`) we need to provide a new allocator that handles: -...
## The Problem Currently logging in coreneuron is done through a variety of `printf`s, `DEBUG` ifdefs and a bit of MPI to print only on rank 0. This code is...
Could the nrn_mpi wrappers be a little more high-level and accept vectors and hide the data-unwrapping `.data()` inside? Are they used inside modfile verbatims? If so then this is probably...
@alkino had also mentioned this. Currently, NEURON/CoreNEURON mechanisms access functions and data from CorNEURON without any clear seperation between the codes and no clearly defined interface. We should define what...
This is a leftover from #229 that we left for later. `nrniv_decl.h` collects a number of corenuron declarations from various compilation units but it is not really used. Instead we...
- Protect aligned_alloc from potentially invalid 0 alignment argument - Handle alignment = 0 as normal malloc - Allocate MembList and ThreadMembList objects without alignment. They are small objects and...
**Describe the issue** Following the developer build workflow on macos (intel) with gcc-12 and otherwise default flags, I get the test `queuing_test` failing with following message: ``` 4: unknown location:0:...