fixed a bug

fixed a bug

fixed a bug in testing

made it compile

made changes to debug the last version

Yen-Chen Chen authored 1 month ago and Pradipta Samanta committed 1 month ago

## What is the bug
The `nvhpc_gpu` manual test creates a new pipeline.
## How do you fix it
The manual job was created twice: once upon push and once by merge request event. Therefore, the second pipeline appears. Disabling the merge request event job fixes the CI duplication.

Approved-by: Pradipta Samanta <samanta@dkrz.de>
Merged-by: Pradipta Samanta <samanta@dkrz.de>
Changelog: bugfix

Pradipta Samanta authored 1 month ago and Yen-Chen Chen committed 1 month ago

Enabled the use of different partitions for CI jobs running on CPU and GPU (!47)

## What is the bug
CI jobs always ran on the shared partition, irrespective of whether they were on CPUs or GPUs
## How do you fix it
moved the place of defining the 'SCHEDULAR_PARAMETERS' 

Approved-by: Yen-Chen Chen <yen-chen.chen@tum.de>
Merged-by: Yen-Chen Chen <yen-chen.chen@tum.de>
Changelog: bugfix

Yen-Chen Chen authored 1 month ago and Pradipta Samanta committed 1 month ago

## What is the bug
1. CMake warning pops up when one uses FetchContent without setting the download extract time stamp or CMake Policy 0135.
2. `libfortran-support` has been made public, but we are still fetching it via ssh.
## How do you fix it
Specify CMake Policy 0135, same as [here in test/fortran](https://gitlab.dkrz.de/icon-libraries/libiconmath/-/blob/main/test/fortran/CMakeLists.txt?ref_type=heads#L13-15

), fetch `libfortran-support` with https.

Merged-by: Pradipta Samanta <samanta@dkrz.de>
Changelog: bugfix

Dylan Kierans authored 1 month ago and Pradipta Samanta committed 1 month ago

## What is the bug
This MR:
* Reverts all variables in `set_zlev()` back from double-precision to working-precision.
* Adds a single & double precision variants (previously working-precision) of `rotate_latlon()`

## Details
Reverting `set_zlev()` changes I introduced in !13

. This will only effect single-precision builds.

Co-authored-by: Pradipta Samanta <samanta@dkrz.de>
Merged-by: Pradipta Samanta <samanta@dkrz.de>
Changelog: feature

Pradipta Samanta authored 1 month ago and Yen-Chen Chen committed 1 month ago

## What is the bug
The CMake option `IM_ENABLE_OPENACC` was by mistake written as `MS_ENABLE_OPENACC` in `gitlab-ci.yml`
## How do you fix it
Using the correct option now

Approved-by: Yen-Chen Chen <yen-chen.chen@tum.de>
Merged-by: Yen-Chen Chen <yen-chen.chen@tum.de>
Changelog: bugfix

Pradipta Samanta authored 2 months ago and Yen-Chen Chen committed 2 months ago

## What is the bug
The use of `c_int` as a variable name in the module `mo_lib_interpolation_scalar` would create conflicts during the C++ implementation of the code. This MR solves the issue #6


## How do you fix it
The variable name `c_int` is changed to `coeff_int`

Approved-by: Yen-Chen Chen <yen-chen.chen@tum.de>
Merged-by: Yen-Chen Chen <yen-chen.chen@tum.de>
Changelog: bugfix

Pradipta Samanta authored 2 months ago and Yen-Chen Chen committed 2 months ago

## What is the bug
The `config.cmake.in` was missing `find_dependency(fortran-support)`
## How do you fix it
Added the missing call of `find_dependency`. Also, made the call more general and only for packages which are previously found using `find_package` during the compilation of `libiconmath`

Merged-by: Yen-Chen Chen <yen-chen.chen@tum.de>
Changelog: bugfix

Yen-Chen Chen authored 2 months ago and Pradipta Samanta committed 2 months ago

## What is the bug
Library links should be public. See #3. This closes #3

.
## How do you fix it
Change `PRIVATE` to `PUBLIC`.

Approved-by: Pradipta Samanta <samanta@dkrz.de>
Merged-by: Pradipta Samanta <samanta@dkrz.de>
Changelog: bugfix

Yen-Chen Chen authored 2 months ago and Pradipta Samanta committed 2 months ago

## What is the new feature
Update the version of `libfortran-support` to the latest, which also supports single precision.

Merged-by: Pradipta Samanta <samanta@dkrz.de>
Changelog: feature

Dylan Kierans authored 2 months ago and Pradipta Samanta committed 2 months ago

## What is the new feature
Support for single-precision as working-precision in libiconmath. Related to fortran-support (libfortran-support!98)

Related to icon-mpim milestone (https://gitlab.dkrz.de/icon/icon-mpim/-/milestones/14#tab-issues)

## How is it implemented
- Precision handling: Dependent on fortran-support `src/mo_iconlib_kind.f90`
- Extensions of some functions to include sp and dp interfaces (independent of wp/sp).
- Related branches before libiconmath merger:
  - libmath-interpolation/feature-single-precision (https://gitlab.dkrz.de/icon-libraries/libmath-interpolation/-/tree/feature-single-precision?ref_type=heads)
  - libmath-support/feature-single-precision (https://gitlab.dkrz.de/icon-libraries/libmath-support/-/tree/feature-single-precision?ref_type=heads

)

Co-authored-by: Pradipta Samanta <samanta@dkrz.de>
Co-authored-by: Dylan Kierans <dylankierans53@gmail.com>
Merged-by: Pradipta Samanta <samanta@dkrz.de>
Changelog: feature

Yen-Chen Chen authored 2 months ago and Pradipta Samanta committed 2 months ago

## What is the new feature
Bump license year to 2025

Merged-by: Pradipta Samanta <samanta@dkrz.de>
Changelog: other

Daniel Reinert authored 3 months ago and Pradipta Samanta committed 3 months ago

Generalize field declaration for the least squares reconstruction routines (!24)

## What is the new feature

The field declaration of the LSQ reconstruction routines has been generalized slightly, in order to make the routines applicable to the ocean surface wave model **ICON-wave**. 

## How is it implemented

Currently, the 'vertical' or '3rd' dimension of the local field `z_b` is implicitly identified with `p_patch%nlev`. This is fine for the atmosphere model, but does not work for ICON-WAVE, where the 3rd dimension is the total number of spectral wave energy bins. 
This issue has been solved by replacing 
```
REAL(wp) :: z_d(3, nproma, nlev)
``` 
with
```
REAL(wp) :: z_d(3, nproma, elev)
```
As a side effect, `nlev` is no longer needed by the reconstruction routines. The argument lists have been adapted accordingly.

Merged-by: Pradipta Samanta <samanta@dkrz.de>
Changelog: default

Pradipta Samanta authored 3 months ago and Yen-Chen Chen committed 3 months ago

## What is the bug
OpenMP optimization determines the values of `i_startblk` and `i_endblk` on its own instead of the value passed in the subroutine 
## How do you fix it
Removed the inclusion of `i_startblk` and `i_endblk` as private variables while starting an OMP parallel construct. It was redundant anyway. 

Approved-by: Yen-Chen Chen <yen-chen.chen@kit.edu>
Merged-by: Yen-Chen Chen <yen-chen.chen@kit.edu>
Changelog: bugfix