-
Notifications
You must be signed in to change notification settings - Fork 0
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Porting the tracer advection to gt4py #252
Conversation
Co-authored-by: Erwan Cossevin <[email protected]> Co-authored-by: Stephanie Westerhuis <[email protected]>
* run pre-commit * fix hflx_limiter_pd_stencil_01
* adding hflx_limiter_mo_stencil_02 * changes order of args and return a tuple * Revert "changes order of args and return a tuple" This reverts commit 99638df. * porting step_advection_stencil_01 * porting step_advection_stencil_02 * porting upwind_hflux_miura_stencil_02 * hflx_limiter_mo_stencil_04 * fix step_advection_stencil_01 * hflx_limiter_mo_stencil_03 * clean up: move upwind_hflux_miura_stencil_02.py * add test for conditional branch in hflx_limiter_mo_stencil_04 * switch to using min_over in hflx_limiter_mo_stencil_03 * refactor hflx_limiter_mo_stencil_04, fix tests for Koffset * add upwind_vflux_ppm_stencil_01 * upwind_hflux_miura_stencil_02 - fix sparse field issue, split into separate sums instead of using neighbor_sum * remove commented out code * fix: upwind_vflux_ppm_stencil_01 * fix: hflx_limiter_mo_stencil_01: return tuple, use int32 * hflx_limiter_pd_stencil_02.py: use int fields
Co-authored-by: Magdalena Luz <[email protected]> Co-authored-by: Nicoletta Farabullini <[email protected]>
* fixing return type * removing broadcast
(feat) - stencil to set field to zero for Field[[CellDim], float] and Field[[CellDim, KDim], float]
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
It looks good to me. Just make sure the new changes from main fixing pre-commit config are incorporated in the PR before merging.
…cils Conflicts: model/atmosphere/dycore/src/icon4py/model/atmosphere/dycore/mo_velocity_advection_stencil_18.py model/atmosphere/dycore/src/icon4py/model/atmosphere/dycore/mo_velocity_advection_stencil_20.py model/common/src/icon4py/model/common/test_utils/simple_mesh.py tools/pyproject.toml
launch jenkins spack |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Looks good!
launch jenkins spack |
merging of greenline with main
* add option to use custom spack-c2sm branch for spack tests * fix missing declarations * fix missing declarations --------- Co-authored-by: Jonas Jucker <[email protected]>
This fixes test references where the current implementation assumes a certain behavior of the backend. Specifically, we used np.roll in the references and the backend (by chance) had the same wrap-around behavior. For an upcoming PR in GT4Py, we need to remove that expectation. --------- Co-authored-by: Jonas Jucker <[email protected]> Co-authored-by: juckerj <[email protected]> Co-authored-by: Abishek Gopal <[email protected]>
Fix static typing annotations in icon4pytools.
This PR allows to use the DELETE statements only for the fused stencil mode. --------- Co-authored-by: Samuel <[email protected]>
57852e5
to
d4436c0
Compare
…cils Conflicts: .github/workflows/icon4py-qa.yml model/README.md model/common/src/icon4py/model/common/test_utils/simple_mesh.py model/requirements-dev.txt model/requirements.txt model/tox.ini requirements.txt tools/requirements-dev.txt tox.ini
launch jenkins spack |
launch jenkins spack spackProject=C2SM/pyicon4py_with_advection |
1 similar comment
launch jenkins spack spackProject=C2SM/pyicon4py_with_advection |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
ILGTM. Kudos for the great work!
Mandatory Tests Please make sure you run these tests via comment before you merge!
Optional Tests In case your change might affect downstream icon-exclaim, please consider running
For more detailed information please look at CI in the EXCLAIM universe. |
launch jenkins spack spackProject=C2SM/pyicon4py_with_advection |
Porting and integrating the stencils that are in the operational code path of the tracer advection according to the namelist:
Exceptions: 3
DO WHILE
loops inmo_advection_vflux.f90
couldn't be ported.TODO when
ASYNC(1)
are added toadvection/mo_advection_quadrature.f90
:TODO when multiple stencils can be used with different
KDim_size
inCMakelists.txt
:dsl_advection_cycling
Remark:
__init__.py
files have been deleted in the tests folders indycore
andadvection
. Otherwisepytest
inmodel/
wouldn't work. Another alternative would have been to rename the tests folderstests_dycore/
andtests_advection/
. This is still to be discussed. EDIT: Changing the names of the folders to<components>_tests
as in https://github.com/C2SM/icon4py/pull/260/files#diff-8bce9783214998f11bc562a6e5619e3f25287303b4d210afe4ed6115f80b1c05 has been chosen (see comments)