Started by an SCM change177016149945a63eeed7527e8b97673c335ca5e3f6f016149945a63eeed7527e8b97673c335ca5e3f6frefs/remotes/origin/clubb_silhs_devel016149945a63eeed7527e8b97673c335ca5e3f6f016149945a63eeed7527e8b97673c335ca5e3f6frefs/remotes/origin/clubb_silhs_devel016149945a63eeed7527e8b97673c335ca5e3f6f016149945a63eeed7527e8b97673c335ca5e3f6frefs/remotes/origin/clubb_silhs_develhttps://github.com/larson-group/cam.githttp://carson.math.uwm.edu/jenkins/blue/organizations/jenkins/cam_gpu_test/detail/cam_gpu_test/177/artifactshttp://carson.math.uwm.edu/jenkins/blue/organizations/jenkins/cam_gpu_test/detail/cam_gpu_test/177/changeshttp://carson.math.uwm.edu/jenkins/blue/organizations/jenkins/cam_gpu_test/detail/cam_gpu_test/177/http://carson.math.uwm.edu/jenkins/blue/organizations/jenkins/cam_gpu_test/detail/cam_gpu_test/177/testsfalseCheckout Externals and Copy Custom FilesRemove Old OutputRunning ne3 with OpenACCfalse#17747855037162916cam_gpu_test #177177false1777538SUCCESS1787645766383http://carson.math.uwm.edu/jenkins/view/Production%20View/job/cam_gpu_test/177/advance_clubb_core_module.F90a593094240550d7eb4287522b28dc4fc28beae7b1787604006000http://carson.math.uwm.edu/jenkins/user/noreplynoreplynoreply@github.comRemove a blank line in order to trigger nightly tests Removed unnecessary blank line at the beginning of the module. 2026-08-24 15:40:06 -0500a593094240550d7eb4287522b28dc4fc28beae7bRemove a blank line in order to trigger nightly testseditadvance_clubb_core_module.F90githttp://carson.math.uwm.edu/jenkins/user/noreplynoreplynoreplyfalseStarted by an SCM change178cd8186b311dcb841754fc4bd8979ec1ca7b12679cd8186b311dcb841754fc4bd8979ec1ca7b12679refs/remotes/origin/clubb_silhs_develcd8186b311dcb841754fc4bd8979ec1ca7b12679cd8186b311dcb841754fc4bd8979ec1ca7b12679refs/remotes/origin/clubb_silhs_develcd8186b311dcb841754fc4bd8979ec1ca7b12679cd8186b311dcb841754fc4bd8979ec1ca7b12679refs/remotes/origin/clubb_silhs_develhttps://github.com/larson-group/cam.gitfalse#178139719127162916cam_gpu_test #178178false1787898SUCCESS1788423365783http://carson.math.uwm.edu/jenkins/view/Production%20View/job/cam_gpu_test/178/stats_netcdf.F90faeef517653ebf7aeab4fc52be8f5019101dd0621788372901000http://carson.math.uwm.edu/jenkins/user/noreplynoreplynoreply@github.comClubb JAX (#1351) * Testing new porting shortcut by creating a jax version of advance_xm_wpxp.py - it wen't well. Also updating the porting shortcut with more tips. * Small update to porting shortcut * Iterative improvements to porting shortcut * Using porting script to port advance_xp2_xpyp in one go - still some issues that seems to require rerunning to fix issues, which is sad but maybe the best we can do with such a simple method * Little bug fixes * Updates to make the ports create better jax in the sense that it is more jit-ready * Adding a more general plan to also use CLUBB-JAX and using it on advance_xm_wpxp * Cleanup * More properly jax-izing the current jax ports * More formatting of ported jax code * More formatting of ported jax code * Updates in pursuit of full jit-compatibility * Adding jit compile test * Fully jitting advance_xp2_xpyp * Copying parameter_indices from CLUBB-JAX but subtracting 1 to avoid -1 in code * Little update to port shortcut * Jitting and formatting CLUBB-JAX * Mostly formatting * Updating jax code * Improving stats outputting * Compile mono flux range loops with JAX * Improvement to jax mixing length port * Organizing stats code * Making jax work on GPU * Improvements to prescribe forcings code * Mainly prescribe forcings formatting work * Small improvements, getting rid of unneccesary weird jit statement * Radiation porting completed. Bugsrad and silhs radiation were intentionally left out for now. * Removing stale code and adding a proposal.md file detailing a potential future feature to make the python_api work with current jax code * Proper port of model flags and parameters and constants and such - those there were all kinda put into one file before * new jenkinsfile added * Tiny cleanup and adding back some stats outputs for microphysics that we need for some reason * limited to only one process. Find out built time * made jenkins jax tests faster by running them not so extensively * Mainly working on README, also some other tiny things: changing default test stats list to standard instead of all_stats, and renaming run_jax.sh to run_jax_wrapper.sh --------- Co-authored-by: Daniel Ebel <debel@uwm.edu> 2026-09-02 13:15:01 -0500faeef517653ebf7aeab4fc52be8f5019101dd062Clubb JAX (#1351)editstats_netcdf.F90githttp://carson.math.uwm.edu/jenkins/user/noreplynoreplyfalse179http://carson.math.uwm.edu/jenkins/view/Production%20View/job/cam_gpu_test/179/177http://carson.math.uwm.edu/jenkins/view/Production%20View/job/cam_gpu_test/177/Started by an SCM change176057591de860e03dfa87c5b28ab0a38e2b92cec0f057591de860e03dfa87c5b28ab0a38e2b92cec0frefs/remotes/origin/clubb_silhs_devel057591de860e03dfa87c5b28ab0a38e2b92cec0f057591de860e03dfa87c5b28ab0a38e2b92cec0frefs/remotes/origin/clubb_silhs_devel057591de860e03dfa87c5b28ab0a38e2b92cec0f057591de860e03dfa87c5b28ab0a38e2b92cec0frefs/remotes/origin/clubb_silhs_develhttps://github.com/larson-group/cam.gitfalse#17631868067162916cam_gpu_test #176176false1767301SUCCESS1787300166907http://carson.math.uwm.edu/jenkins/view/Production%20View/job/cam_gpu_test/176/setup_clubb_pdf_params.F90pdf_closure_module.F90adg1_adg2_3d_luhar_pdf.F906bea7684fe3a0a67b6554302c5aaca5b83a86a011787260825000http://carson.math.uwm.edu/jenkins/user/noreplynoreplynoreply@github.comImprove Dash workflows and broker services (#1342) * Improve Dash workflows and broker services Streamline run telemetry and output-scoped admission, add build selection and pyplotgen export support, improve profile and tuning workflows, and make Tune failures column-local and recoverable. * A little cleanup * Simplification in SILHS to keep errors silent in debug=-1 mode, and removing accidental change in advance_clubb_core.F90 * Little dash fix and comment improvement * Replacing scipy usage in the adam initial point choosing with numpy * Removing l_rtm_nudge from flag set file used by run_bindiff_w_flags tests. This is technically a model_setting, so it shouldn't be here, and is already enabled by cgils_s12, so will remain tested in at least one case. 2026-08-20 16:20:25 -05006bea7684fe3a0a67b6554302c5aaca5b83a86a01Improve Dash workflows and broker services (#1342)editpdf_closure_module.F90editadg1_adg2_3d_luhar_pdf.F90editsetup_clubb_pdf_params.F90silhs_importance_sample_module.F90latin_hypercube_driver_module.F9056d859f2e74f340304ae3bf48a46cbf0ae26300c1787260825000http://carson.math.uwm.edu/jenkins/user/noreplynoreplynoreply@github.comImprove Dash workflows and broker services (#1342) * Improve Dash workflows and broker services Streamline run telemetry and output-scoped admission, add build selection and pyplotgen export support, improve profile and tuning workflows, and make Tune failures column-local and recoverable. * A little cleanup * Simplification in SILHS to keep errors silent in debug=-1 mode, and removing accidental change in advance_clubb_core.F90 * Little dash fix and comment improvement * Replacing scipy usage in the adam initial point choosing with numpy * Removing l_rtm_nudge from flag set file used by run_bindiff_w_flags tests. This is technically a model_setting, so it shouldn't be here, and is already enabled by cgils_s12, so will remain tested in at least one case. 2026-08-20 16:20:25 -050056d859f2e74f340304ae3bf48a46cbf0ae26300cImprove Dash workflows and broker services (#1342)editsilhs_importance_sample_module.F90editlatin_hypercube_driver_module.F90githttp://carson.math.uwm.edu/jenkins/user/noreplynoreplyfalse177http://carson.math.uwm.edu/jenkins/view/Production%20View/job/cam_gpu_test/177/175http://carson.math.uwm.edu/jenkins/view/Production%20View/job/cam_gpu_test/175/