Hi all,
Einstein Toolkit Meeting - August 31, 2023
Chair : Peter Minute taker : Leo Present : Sam Cupp, Yosef Zlochower, Peter Diener, Leo Werneck, Zach Etienne, Roland Haas
* Next ET Release
- Should CarpetX be enabled or disabled by default? Sam believes it would be weird to include all of the new thorns that depend on CarpetX and then have people do extra work to have access to them. This will force everyone to have a compiler that supports C++17, but this is not a particularly high bar. Erik also supports enabling it by default.
- Roland said some modification might be needed for simulation factory to account for this. There is also some consideration about the default compiler, which would mostly affect the "generic" configuration. The new configuration would need to check the system until it finds a C++17-compatible compiler, something that is currently not done. Further, some older versions of GCC require you to pass additional options to link against the correct, "experimental" standard library.
- Zach shared a useful list of supported C++ features by compiler: https://en.cppreference.com/w/cpp/compiler_support/17
- Roland suggested enabling CarpetX now so that problems can be fixed as they appear.
- Loop control, as it exists, prevents Carpet and CarpetX from existing in the same executable. The provided macros are drastically different for the two drivers because of the way looping is done in AMReX. One option is to have thorns require "Loop control" OR "Loop control X" so that the header file can pick the appropriate macros at compile time. This is not yet implemented, but Roland will work on it.
- Peter should review GRHayLHDX, but skip GRHayLMHDX.
* Unanswered Questions on the Mailing List
- There are bugs in POWER. Roland will look into it.
* Open Tickets:
- #2754: there are different definitions of magnetic fields in the toolkit, differing by a factor of sqrt(4pi). Note that the evolution of GRMHD quantities is still consistent, as these factors are accounted for when computing the magnetic pressure that is used to compute the energy-momentum tensor. As Zach points out, this only affects very specific scenarios where direct comparisons between e.g., GRHydro and IllinoisGRMHD are performed.
- In the future, it would be nice do modify the Seed_Magnetic_Fields thorns from WVUThorns* to provide a more user friendly way of setting up the magnetic fields (for example, the user can specify the ratio of magnetic to fluid pressure).
Next week's chair : Zach Next week's minute taker : Roland
Cheers, Leo
------ Leonardo R. Werneck, Ph.D. Postdoctoral researcher Office EP 314 | Department of Physics | University of Idaho 875 Perimeter Dr. MS 0903 Moscow, ID 83844-0903, USA leonardo@uidaho.edu mailto:leonardo@uidaho.edu https://leowerneck.github.io https://leowerneck.github.io/