Merge 'regional_boundary_update' from upstream #40
Conversation
…all that caused a reproducibility issue.
Modify the halo extents of u and vt in the regional_boundary_update call
…all that caused a reproducibility issue. (#55) Co-authored-by: Chan-hoo.Jeon <chan-hoo.jeon@noaa.gov>
|
Someone will need to pull this into an upcoming PR. I suspect it will change the results of all regional tests. |
|
@DusanJovic-NOAA can we just pull the entire noaa-gfdl dev/emc down to noaa-emc? In addition to the one commit shown here, there are documentation updates in |
merge ufs-release/public-v2 -> dev/emc
Revert "merge ufs-release/public-v2 -> dev/emc"
|
I see only differences in model/fv_regional_bc.F90 when I try to open PR. |
|
This is weird.
I just changed the submodule pointer from NOAA-EMC dev/emc to NOAA-GFDL dev/emc and I am seeing the changes here:
NOAA-EMC/fv3atm#178 <NOAA-EMC/fv3atm#178>
… On Sep 28, 2020, at 10:04 AM, Dusan Jovic ***@***.***> wrote:
I see only differences in model/fv_regional_bc.F90 when I try to open PR.
—
You are receiving this because you commented.
Reply to this email directly, view it on GitHub <#40 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AB5C2ROTZKC3APB7W4SS3KTSICXZ7ANCNFSM4RZSLMWQ>.
|
|
In the future all changes should be made to NOAA_EMC repo first. If they go directly to NOAA_GFDL (like documentation and this bug fix in fv_regional_bc.F90) then there's no need for us to maintain this fork, we can just point to GFDL's repo directly from UFS, like we do with ccpp, for example. |
|
Can we merge this PR? |
This should be part of the regression testing conducted in my upcoming ufs-weather-model PR. The fv3atm PR handles the submodule pointer update for GFDL_atmos_cubed_sphere. If you can find a way to update this PR so that it shows the documentation updates, too, then the code I am testing in NOAA-EMC/fv3atm#178 will be identical to your PR. |
|
Associated PRs: #40 For regression testing information, see ufs-community/ufs-weather-model#208. |
|
Update: after some back and forth, we know that this PR is correct and sufficient. The "missing" documentation changes were merged into NOAA-EMC dev/emc previously without updating the submodule pointer in fv3atm or running the regression tests. This PR #40 will make NOAA-GFDL dev/emc and NOAA-EMC dev/emc identical and should be merged after the testing as outlined in ufs-community/ufs-weather-model#208 is complete. |
|
@climbfuji That is what I think. |
There was a problem hiding this comment.
This PR will be committed along with ufs-weather-model PR NOAA-GFDL#208
No description provided.