-
Notifications
You must be signed in to change notification settings - Fork 15
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
Error cleanup (Intel Fortran) #130
Conversation
This patch clears out many errors detected by Intel Fortran. Most are false positives from stub functions which would normally be replaced in production and report unset output. These variables are now assigned dummy values in order to pacify the compiler. The `stat` function in POSIX was incorrectly passing its `buf` object to the C `stat` function as `intent(in)`, causing the compiler to believe that the contents were unset. Oddly, this was already working correctly, and perhaps warrants further investigation, but it has now been correctly set to `intent(inout)`. The `ppoly_*` variables in `check_reconstruction_1d` appear to have been incorrectly declared as `out`, when they are clearly used as `in` to validate the values. This has been corrected. `register_diag_field` in the ice shelf diag manager was incorrectly declared and the function appeared to return nothing. Perhaps this function was not used for anything. An IO statement in MOM_open_boundary had a syntax error; this has been fixed. `get_dataset` returns a `dataset_type`, so some compilers expect the stub function to also return a valid `dataset`. Since the stub `dataset_type` contains no fields, any locally declared instance should be sufficient as a return value.
Codecov ReportAttention: Patch coverage is
Additional details and impacted files@@ Coverage Diff @@
## dev/emc #130 +/- ##
===========================================
- Coverage 37.45% 37.44% -0.02%
===========================================
Files 270 270
Lines 79765 79800 +35
Branches 14833 14833
===========================================
Hits 29878 29878
- Misses 44350 44385 +35
Partials 5537 5537 ☔ View full report in Codecov by Sentry. |
@jiandewang @DeniseWorthen are you able to provide a review for this sub-component PR? UFS-WM PR 2197 could be an option to test against soon. |
thanks for @marshallward work. It eliminated those warning now. |
@jiandewang testing is complete on UFS-WM PR #2202. Please merge this MOM6 PR when you're able. |
just merged |
This patch clears out many errors detected by Intel Fortran.
Most are false positives from stub functions which would normally be replaced in production and report unset output. These variables are now assigned dummy values in order to pacify the compiler.
The
stat
function in POSIX was incorrectly passing itsbuf
object to the Cstat
function asintent(in)
, causing the compiler to believe that the contents were unset. Oddly, this was already working correctly, and perhaps warrants further investigation, but it has now been correctly set tointent(inout)
.The
ppoly_*
variables incheck_reconstruction_1d
appear to have been incorrectly declared asout
, when they are clearly used asin
to validate the values. This has been corrected.register_diag_field
in the ice shelf diag manager was incorrectly declared and the function appeared to return nothing. Perhaps this function was not used for anything.An IO statement in MOM_open_boundary had a syntax error; this has been fixed.
get_dataset
returns adataset_type
, so some compilers expect the stub function to also return a validdataset
. Since the stubdataset_type
contains no fields, any locally declared instance should be sufficient as a return value.