Ian Henriksen

Weirdly enough, it's now segfaulting at interpreter startup for me. Once again, I'm not sure what's changed. I haven't rebuilt glibc.

We're doing enough crazy stuff in the VEC code to have weird stuff happen, but somehow it continues to surprise me.

Okay, at lest in the case of numba, the issue appears to actually be bad handling of RTLD_LAZY. (Use-case is here: https://github.com/numba/numba/blob/145e4435c9bea78071634db6084d8329129748ab/numba/__init__.py#L152). Hopefully this is just an issue with our...

Not sure what was wrong with the old MKL loading logic though. That was with RTLD_NOW set.

Maybe that's it. An quick reading earlier didn't show anything like that, but the log is showing RTLD_NOLOAD instead of RTLD_GLOBAL so something's off there.

Yah, there's nothing obviously wrong in https://github.com/ut-parla/Parla.py/blob/master/runtime_libs/virt_dlopen.c#L30. No idea where the RTLD_NOLOAD is coming from, but it's before that line.

Okay, there's a lot going on here and I'm getting a bit disoriented. Here's what I've managed to confirm: - Numba's try/except doesn't normally find the library in my current...

Maybe `execution_contexts` or `virtual_contexts` or something like that. `VECs` seems a bit brief, but spelling it out fully seems too verbose.

@arthurp mentioned in https://github.com/ut-parla/Parla.py/issues/39#issuecomment-830121508 that this will require some changes to both the device model and scheduler. I agree. @arthurp if you have any additional thoughts beyond this it'd be...

Yah, my only opinion on this is that it shouldn't be doing what it does currently. We need to either disallow the nonlocal and raise an error or allow it...