+New config_src directory structure#1339
Conversation
- this addresses a bug in the apply_ALE_sponge routine which had been passing zero values to the remapping routine for the target grid thicknesses.
…_022321 Ale sponge fix 022321
…ate-2021-02-19 Dev gfdl main candidate 2021-02-19
Merge main to dev/gfdl
Created the new config_src directory tree structure, as agreed upon at https://github.com/NOAA-GFDL/MOM6/discussions/1286, to eventually accomodate the selection of different infrastructures. No .F90 files are changed but there are small changes to ac/configure.ac to accomodate the new structure while also allowing for the old target to use the old structure. All answers are bitwise identical, and all MOM6-examples and TC tests are passing.
Codecov Report
@@ Coverage Diff @@
## gfdl-fms2 #1339 +/- ##
==========================================
Coverage 45.74% 45.75%
==========================================
Files 234 234
Lines 72540 72542 +2
==========================================
+ Hits 33187 33189 +2
Misses 39353 39353
Continue to review full report at Codecov.
|
| # a compile-time mode, so this is not exactly being used as intended. | ||
| MEM_LAYOUT=${srcdir}/config_src/dynamic_symmetric | ||
| MEM_LAYOUT=${srcdir}/config_src/memory/dynamic_symmetric | ||
| AC_CHECK_FILE($MEM_LAYOUT, [MEM_LAYOUT=$MEM_LAYOUT], [MEM_LAYOUT=${srcdir}/config_src/dynamic_symmetric]) |
There was a problem hiding this comment.
I'm guessing this directory also needs to be updated.
There was a problem hiding this comment.
No, it is correct as written. The point here is that if the MEM_LAYOUT file does not exist, it is because this is the target directory that is still using the old directory structure. Because I know essentially nothing about autoconf, I spent much more time this morning figuring out how to make the TC make commands work with both the old and new directory structures than I did with actually moving the files around!
There was a problem hiding this comment.
But it's looking in defaulting to a directory which ought to no longer exist. Is this being done to accommodate the regression test?
There was a problem hiding this comment.
Without this line the TC regression tests were failing to compile; with it they do. Once we are no longer interested in comparing runs with potentially different directory structures, we can get rid of the ACC_CHECK_FILE line.
There was a problem hiding this comment.
Then at the least there needs to be a comment explaining why this is here. It's a nonsense macro in the context of the repository. While we do provide a regression test, we should not leave around musty old content to accommodate it.
There was a problem hiding this comment.
As a point of reference, the very last PR to main also did not pass regression.
Updated the MOM6/docs files and .gitlab-ci.yml to reflect the new config_src directory structure.
3 participants
Created the new config_src directory tree structure, as agreed upon at
https://github.com/NOAA-GFDL/MOM6/discussions/1286, to eventually accommodate
the selection of different infrastructures. No .F90 files are changed but there
are small changes to ac/configure.ac to accommodate the new structure while also
allowing for the old target to use the old structure. All answers are bitwise
identical, and all MOM6-examples and TC tests are passing.
This change will require modifications to various compile scripts.