#2944: CCE_Export Issues Ticket
Reporter: Zach Etienne
Status: new
Milestone:
Version:
Type: bug
Priority: major
Component:
Comment (by Roland Haas):
Clearly we need a policy on wordy LLM generated reviews. Ideally one ticket per issue would be used, making progress easier to track.
--
Ticket URL: https://bitbucket.org/einsteintoolkit/tickets/issues/2944/cce_export-issues…
#2922: CarpetX does not compile with CUDA 12.9
Reporter: Steven R. Brandt
Status: resolved
Milestone: ET_2026_05
Version:
Type: bug
Priority: major
Component: CarpetX
Changes (by Roland Haas):
status: resolved (was new)
--
Ticket URL: https://bitbucket.org/einsteintoolkit/tickets/issues/2922/carpetx-does-not-…
#2922: CarpetX does not compile with CUDA 12.9
Reporter: Steven R. Brandt
Status: new
Milestone: ET_2026_05
Version:
Type: bug
Priority: major
Component: CarpetX
Comment (by Roland Haas):
This seems to actually have been due to newer compilers pulling in C++23 stdc++ library files, which push more names into the `std` namespace. Pull request https://github.com/EinsteinToolkit/CarpetX/pull/381 addresses this and resolves this ticket as well.
--
Ticket URL: https://bitbucket.org/einsteintoolkit/tickets/issues/2922/carpetx-does-not-…
#2949: use c99 bool instead of typedefing our own
Reporter: Roland Haas
Status: new
Milestone:
Version:
Type: bug
Priority: minor
Component:
C23 makes `bool` a reserved keyword and provides the type by default making some thorns (and the flesh) fail to compiled.
These three pull requests add the required `#include <stdbool.h>` to the respective files. Since this exists in c99, which Cactus requires, this is fine.
--
Ticket URL: https://bitbucket.org/einsteintoolkit/tickets/issues/2949/use-c99-bool-inst…
#2174: Test "Multi Patch Scalar Wave Equation" example
Reporter: Roland Haas
Status: resolved
Milestone: ET_2026_05
Version: development version
Type: task
Priority: major
Component: EinsteinToolkit website
Changes (by Lucas Timotheo Sanches):
status: resolved (was open)
Comment (by Lucas Timotheo Sanches):
Updated gallery example for ET_2026_05
--
Ticket URL: https://bitbucket.org/einsteintoolkit/tickets/issues/2174/test-multi-patch-…
#2948: HydroBaseX: Missing HydroBase features (Aphi GF, conditional storage, split init)
Reporter: Maxwell Rizzo
Status: new
Milestone:
Version:
Type: bug
Priority: minor
Component: CarpetX
Current HydroBaseX has a single routine that zeros all the variables it declares, along with a single parameter controlling whether this routine runs: `initial_hydro`. This parameter was used/extended by initial data thorns when they provide hydro quantities.
Following the legacy Carpet/HydroBase convention, `initial_hydro` should only control the minimal hydro fields, with other GFs accompanied by corresponding `initial_*` parameters and routines. The current implementation works, but is awkward for initial hydro data thorns to integrate with: they can either avoid using `initial_hydro` themselves and let HydroBaseX zero all fields and place their data ontop of that, or they can manually zero the fields they are not providing, which is unnecessary when HydroBaseX should be able to do it.
Additionally, HydroBaseX is missing the `Aphi` GF. This is the electric potential, and adding it would round out the EM support alongside `Bvec` and `Avec`.
Lastly, modularizing this initialization should allow conditional storage to be restored, storing optional hydro fields only when the corresponding `initial_*` parameter is not `"none"`, following HydroBase. This should also provide a minor memory/performance improvement.
Addressed by already open PR: [#385](https://github.com/EinsteinToolkit/CarpetX/pull/385)
--
Ticket URL: https://bitbucket.org/einsteintoolkit/tickets/issues/2948/hydrobasex-missin…
#2947: CarpetX: Support 2D output
Reporter: Maxwell Rizzo
Status: new
Milestone:
Version:
Type: enhancement
Priority: minor
Component:
Changes (by Maxwell Rizzo):
Currently 1D slices (x/y/z) and the full 3D output for Grid Functions are both supported output configurations from CarpetX. CarpetX is missing functionality to output GFs sliced along 2D planes (xy, xz, yz).
There is a sublety for how the slice is chosen, given the different centering options, some variables likely will not contain data along the `x/y/z=0` planes exactly, which is a similar sublety that the 1D output slice must resolve, hopefully these can be resolved consistently.
Also from a discussion with Roland, 2D output is inefficient for a Carpet/CarpetX simulation to output, compared to 1D and 3D. A performance test could be useful in gauging if this feature would be worth implementing/using at all in CarpetX.
TSV (or another text file format) probably makes sense as the starting point, if functional a compressed format like OpenPMD would also be desirable, bringing the 2D CarpetX output features in-line to what was present in Carpet.
--
Ticket URL: https://bitbucket.org/einsteintoolkit/tickets/issues/2947/carpetx-support-2…