7d611c56ad
docs(rebuilt-docs): rebuild docs for v0.7.6rc4.0
2026-04-20 12:45:31 -04:00
cf26983fe2
docs(version): version bump v0.7.6rc3.4 -> v0.7.6rc4.0
...
This bump includes all changes addressing ref comments
2026-04-20 12:43:52 -04:00
e704d5a1a7
feat(GridFire): updated
2026-04-20 12:42:23 -04:00
54d3ec9920
docs(gitignore): updated gitignore
2026-04-20 12:41:42 -04:00
b23f5a98c5
feat(examples): added FiPy example
2026-04-20 12:41:27 -04:00
bbd702904a
feat(validation): added more of the scripts to make paper figures
2026-04-20 12:41:10 -04:00
3a22792fd1
fix(GridFire): changes based on ref report
2026-04-20 12:37:53 -04:00
f4d988fa25
feat(benchmarks): added memory and timing benchmarks
2026-04-15 08:23:34 -04:00
ec93720fa0
docs(gitignore): added trace and conanfile
2026-04-13 07:22:51 -04:00
5a1a904e71
refactor(GridFire): updated outputs
2026-04-13 07:19:18 -04:00
d1872cb65a
docs(stubs): updated stubs
2026-04-13 07:18:41 -04:00
c311e4afbd
test(vv): Added more scripts to verify GridFire behavior
2026-04-13 07:18:08 -04:00
84ff182717
feat(GridFire): Added a number of python hooks
...
python hooks to make getting base composition more reliable; further, a number of small changes made to aid in my analysis in response to ref report 1
2026-04-13 07:17:14 -04:00
65297852e5
feat(GF-Version): added auto version header
...
when building the version number is automatically injected into a header now. This allows for more certainty as to what GF version is being used. Note that this is disabled when building the python wheel as there is no clear way to map this dynamically generated header into the wheel structure. This is however not an issue as the python module has a seperate __version__ variable.
2026-04-09 07:45:00 -04:00
45af511db2
fix(Config-Py-Bindings): CVODESolver->PointSolver in python bindings
...
CVODESolver was renamed to PointSolver in the C++ source; however, the python source had not been updated. This has not been made consistent
BREAKING CHANGE:
2026-04-09 07:42:28 -04:00
7ab0d21ba1
fix(cpp_args): properly propegate args
...
previously Cpp args were set with add_project_argument, which does not work from subprojects. Changed this so args can be set from calling projects
2026-02-03 11:20:21 -05:00
917d416ce2
fix(build-check): ! -> not
...
fix broken check for subproject
2026-02-03 08:11:23 -05:00
1cdd2d92e0
fix(degfined_engine_view): removed broken logs
...
some old log statements reference a member no longer present, removed these
2026-02-03 08:02:09 -05:00
4423d7594c
fix(build-check): only add global arguments when not a subproject
...
This is needd to let gridfire work as a subproject
2026-02-03 07:52:24 -05:00
087926728a
docs(version): v0.7.6rc3 -> v0.7.6rc3.1
...
this version uses the target workaround which we use for the SUNDIALS suite to get meson to recognize it as position indepenednt code to CppAD
2025-12-22 08:23:20 -05:00
df09564c9a
fix(cppad): build static lib for cppad
...
this is needed for gcc build and does not break clang build
2025-12-22 08:22:07 -05:00
0bf3ae625b
docs(meson): version v0.7.5rc3 -> v0.7.6rc3
2025-12-22 08:18:02 -05:00
5c1714410a
fix(engine_defined): fixed gcc build warnings
2025-12-22 08:17:23 -05:00
e6a9d8c5bb
fix(wheels): fixed macos wheel generation to pin meson==1.9.1
2025-12-22 07:09:51 -05:00
e73daf88b3
docs(version): v0.7.4rc3 -> v0.7.5rc4
2025-12-20 16:41:27 -05:00
e197227908
fix(meson): address #13
...
Address regresion in meson 1.10.1 by pinning version to 1.9.1
2025-12-20 16:40:28 -05:00
f1f793f775
Merge pull request #12 from tboudreaux/perf/coloring
...
Perf/coloring
2025-12-20 16:17:08 -05:00
e98c9a4050
feat(python): Python multi-test
...
Python now works with mulit-threading and zones
2025-12-20 16:08:33 -05:00
11a596b75b
feat(python): Python Bindings
...
Python Bindings are working again
2025-12-20 16:02:52 -05:00
d65c237b26
feat(fortran): Fortran interface can now use multi-zone
...
Fortran interface uses the new C api ability to call the naieve
multi-zone solver. This allows fortran calling code to make use of in
build parellaism for solving multiple zones
2025-12-19 09:58:47 -05:00
2a9649a72e
feat(C-API): C API now can use multi-zone solver
...
C api has been brought back up to support and can use paraellization along with multi zone solver
2025-12-18 15:14:47 -05:00
cd5e42b69a
feat(omp): useful omp macros
...
A few macros which make turning on and off omp features cleaner without
#defines everywherer
2025-12-18 12:48:10 -05:00
dcfd7b60aa
perf(multi): Simple parallel multi zone solver
...
Added a simple parallel multi-zone solver
2025-12-18 12:47:39 -05:00
4e1edfc142
feat(Spectral): Working on Spectral Solver
2025-12-15 12:14:00 -05:00
0b09ed1cb3
feat(SpectralSolver): Spectral Solver now works in a limited fashion
...
Major work on spectral solver, can now evolve up to about a year. At
that point we likely need to impliment repartitioning logic to stabalize
the network or some other scheme based on the jacobian structure
2025-12-12 17:24:53 -05:00
e114c0e240
perf(thread saftey): All Engines are now thread safe
...
Previously engines were not thread safe, a seperate engine would be
needed for every thread. This is no longer the case. This allows for
much more efficient parallel execution
2025-12-12 12:08:47 -05:00
c7574a2f3d
docs(gitignore): added more subprojects source directories
2025-12-10 12:51:17 -05:00
97a7fd05d2
feat(SpectralSolver): Began work on multizone spectral solver
...
The single zone solver we have is too slow for a true high resolution
multi-zone enviroment. Began work on a spectral element method
multi-zone solver
2025-12-10 12:50:35 -05:00
b57ed57166
build(wrap): updated wrap files
2025-12-10 12:49:21 -05:00
411b2246f0
build(unordered-dense): testing faster map implimentation
...
testing the unordered-dense map implimentation
2025-12-10 12:48:20 -05:00
d990089207
build(cppAD): switched to meson wrap for CppAD
...
We now pull in CppAD with the standard wrap system
2025-12-10 12:47:16 -05:00
8cfa067ad0
perf(GridFire)
...
More preformance improvmnets
1. Switch to mimalloc which gave a roughly 10% improvment accross the
board
2. Use much faster compososition caching
3. Reusing work vector
2025-12-07 12:34:12 -05:00
e48b62f231
perf(GraphEngine): more caching
...
Added more agressive caching, regained about another 50% performance gain. Solar model over 10Gyr now takes roughly 400ms (as opposed to 1 second before)
2025-12-06 16:35:35 -05:00
e95881d917
Merge pull request #10 from tboudreaux/perf/openMP
...
Performance + openMP support
2025-12-06 13:52:55 -05:00
67dde830af
perf(openMP): added openMP support
...
Note that currently this actually slows the code down. Spinning up the threads and tearing them down is expensive
2025-12-06 13:48:12 -05:00
4e2b3cb11f
perf(GraphEngine): using caching clawed back ~10% performance
2025-12-06 12:10:43 -05:00
b6f452e74c
feat(libconfig): new version of libconfig
2025-12-06 11:41:57 -05:00
7242c765f3
build(wasm): major progress on gridfire compiling to wasm
2025-12-03 11:38:08 -05:00
d852ee43fe
perf(precomputation): cleaned up allocations
...
recovered about 5% execution time
2025-12-02 13:09:19 -05:00
ed2c1d5816
build(.gitignore): .whl added
...
.whl files added to gitignore so large precompiled wheel folders are not
accidently commited
2025-12-02 10:04:42 -05:00
