Thomas Herault

It's the same problem with distributed runs: parsec_ce is initialized as late as possible, and when we create these taskpool that load up the termination detector, no comm engine is...

I pushed some commit in PR #642 to handle the lack of device more gracefully, both in the test and the runtime system. However, the test still fails in the...

I'm trying it on leconte. ``` mpirun -np 2 --map-by socket hwloc-info --restrict binding package:0 Package L#0 [...] cpuset = 0x0000ffff,0xf00000ff,0xfff00000 complete cpuset = 0x0000ffff,0xf00000ff,0xfff00000 allowed cpuset = 0x0000ffff,0xf00000ff,0xfff00000 nodeset...

I think the 'proper' CMake way to do that is to expose CUDA, HIP, and other features as COMPONENTS for the PaRSEC package, and the user should use ``` find_package(PaRSEC...

We discussed this today, and it's complicated, because we want to support an empty range too. If we use arbitrary functions a(), b(), and i() and write ``` x =...

I'm a bit lost with CI here... Another pair of eyes would help. To summarize what I observe: - When running in shared=OFF / profiling=ON mode, we don't detect CUDA...

Split this PR in two: one for the tester itself and another for the CI/runners

the info object should be a local that is passed explicitly to the `set` functions, and to the `parsec_init` call. the developer should have a simple way of listing the...

@abouteiller can you test this PR on a HIP machine? Is the HIP port still working? I'm testing on CUDA and LevelZero machines.

> Now that #570 has been merged is this PR still necessary ? Not quite: I still need to import the PTG support of DPC++ part.