-
Notifications
You must be signed in to change notification settings - Fork 156
cam6_4_093: Update CAM testing. #1309
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
Conversation
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
As part of my review, I made sure that each ncdata file which resided in the previous version, still resides in the current one. Aside from a couple of duplicates which were removed since the ic_ymd was not 8000 needed, there were no missing files.
<test compset="QPC5" grid="f10_f10_mg37" name="PLB_D_Ln9" testmods="cam/ttrac_loadbal1"> | ||
<machines> | ||
<machine name="izumi" compiler="nag" category="aux_cam"/> | ||
<machine name="izumi" compiler="nag" category="prealpha"/> | ||
</machines> | ||
<options> | ||
<option name="comment" >301 plb</option> | ||
<option name="comment" >phys_loadbalance=1; 5 test tracers; end of year</option> |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
I would suggest that perhaps the load balance tests (or at least one) should be run on the SE dycore as it is the dycore of the future, while the FV dycore has had discussions about being deprecated
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
The FV dycore is the only one that uses the phys_loadbalance
capability. The other dycores all use the version of phys_grid.F90
which Steve implemented to address weak scaling issues and which does not support physics load balancing.
@@ -1853,9 +1855,9 @@ subroutine coefs_init(coefs_file, available_gases, kdist) | |||
minor_limits_gpt_upper | |||
! Send these to RRTMGP as logicals, | |||
! but they have to be read from the netCDF as integers | |||
logical, dimension(:), allocatable :: minor_scales_with_density_lower, & | |||
logical(wl), dimension(:), allocatable :: minor_scales_with_density_lower, & |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Just for curiosity, why does a logical have a type? Isn't it just true/false?
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
RRTMGP uses the logical kind c_bool
from the iso_c_binding
module for interoperability with C. The dummy arguments of type logical in the kdist%load
function use that kind, so the actual args need to use it as well. NAG is the only compiler we use that distinquishes between c_bool
and kind(.true.)
.
Making progress on (but not resolving):
Issue Most converted CESM regression tests are CAM5 - may want to make most CAM7 #258 - Most converted CESM regression tests are CAM5 - may want to make most CAM7
Issue Convert all tests to the mt232 mask. #1280 - Convert all tests to the mt232 mask.
cleanup ncdata entries in namelist_defaults_cam.xml
updates to allow running cam7 physics with nag-7 on izumi.