Manodeep Sinha

Solved by #295. Now Corrfunc `master` branch should compile and run fine on Apple laptops with M1/M2 cpus Sidenote: Optimised kernels being (slowly) implemented under the `arm64` branch.

I did some further reading (can't find the reference - the name looked familiar - so someone in the open-source python ecosystem) and it is possible that the OpenMP issues...

Creating shared libraries (instead of static libraries) did not resolve the issue of multiple OpenMP runtime libraries (libgomp from compiling with GCC and libiomp5 from mkl within numpy). However, setting...

Setting that environment variable with `os.environ` does not work. Though I am not sure I understand why that is, since the documentation says all child processes should inherit the new...

@beckermr This should be fixed now with #277. As long as `Rmax < Lbox/2`, Corrfunc should correctly compute the pairs. Closing the issue but please feel free to reopen if...

While fixing the build failures (#290), I upgraded the minimum python version to py3.7. That means a bunch of the `os.path` etc routines can now be replaced with `Pathlib.Path` routines

@rainwoodman I assume this is no longer relevant and I am closing the issue. Please feel free to reopen ...

Previously, almost all the time is in the histogram update but that's expected since the histogram update is in the innermost loop. Hard to say with the new concept -...

I will check out the hybrid LUT. Though my experience with the linear binning suggests that the bottleneck is not at the bin-index computing. Not sure I follow the smaller...

Ohhh and forgot about the one optimisation that SWIFT uses - projecting the points along the line joining the cell centers and removing all points that cannot be within Rmax....