update rt.conf and compile.sh to provide cmake options and avoid a translation layer#615
update rt.conf and compile.sh to provide cmake options and avoid a translation layer#615aerorahul wants to merge 19 commits into
Conversation
Change-Id: Ie2103eab6d91105f9cd0e9bb069c2d95bef2b6b8
Change-Id: I7d848f4cd91bea74b04d8b43ab1f45a1ace8cda2
Change-Id: I92025fbae25785f84abf7fa1a08bfbca901bffa6
|
and:
|
|
This is going to cause conflicts everytime develop is updated, so I am going to wait till it is this PR's turn to resolve all conflicts. |
Change-Id: I85510b236d2ef25e09e6f4578ac6196b3a7e2de1
Change-Id: Ie09f7f4cf645bc6661bc49d73fec21e51edfb728
Change-Id: I137a0e798eb2273be61046d081fb379dc0aa0a7a
There is no option with |
I see. Then this second if test must be updated to look for APP that includes WW3 (ATMW, S2SW, etc). |
Change-Id: I9619f724e12e36607ea84a009134e9997d9bddf0
done in 0905841. Please take a look again. |
Change-Id: Ide8e962ffc1f1304c042bb79907789442180338f
|
@aerorahul i am not sure but is this PR is for solving the issue that I faced before. The build was trying to add multiple APP in the build. |
@uturuncoglu Yes I believe so. In this PR, the contents of |
|
@aerorahul Thanks for the clarification. Currently I could not access to Orion and I am trying to setup Cheyenne for testing. I'll let you know when I have a change to test it. Thanks again. |
|
@aerorahul any update about this PR? |
|
@aerorahul @junwang-noaa i am testing this PR along with the #611 and I am getting build error like following when I tried to compile DOCN case, any idea. I'll also review the PR in a more detailed way but first I need to test it under HAFS app with data component configurations. |
|
maybe I need to replace |
|
i also need to add -D before CCPP and APP |
|
@aerorahul okay I pass those errors by adding -D and aand make change in 32BIT option. Now, I am getting following error, Do I need to use special version of CDEPS? We are using NOAA-EMC CDEPS fork as a base in HAFS app but I have additional modifications for ERA5 data atmosphere in it. So, I am not expecting any issue with the CDEPS version that I am using. |
|
@uturuncoglu |
|
@aerorahul okay. no problem. let me know when it is ready and I could test it again. |
|
@uturuncoglu I think I resolved the conflicts. I have not tested them yet. |
| fi | ||
|
|
||
| if [[ ${MAKE_OPT^^} =~ "WW3=Y" ]]; then | ||
| if [[ ${MAKE_OPT^^} =~ "-DAPP=ATMW" ]] || [[ ${MAKE_OPT^^} =~ "-DAPP=S2SW" ]]; then |
There was a problem hiding this comment.
I think that this is not flexible that I explain in the following issue #656. I already completed WW3 coupling through the use of CMEPS under HAFS application and created a set of wave coupled configurations. Listing applications names in this level is not flexible and once someone bring another configuration this file needs to be changed.
There was a problem hiding this comment.
Please make a suggestion/PR and we can review.
All the applications that are present in the UFS-weather-model are tested and supported. So if someone brings in an "application", that needs to be supported.
The entire business of adding if-then-else needs to be revisited.
|
#642 absorbed this PR. |
PR Checklist
Description
compile.shreads in theMAKE_OPTwhich are the deprecated GNU-style options for compilations fromrt.conf.This PR:
rt.confincmakestyle.compile.shto pass throughMAKE_OPTas is tobuild.shwhich callscmake.compile.shandbuild.sh. E.g.SUITESis (was) input toMAKE_OPTandcompile.shwould transform that to-DCCPP_SUITESfor use inbuild.sh. Hereafter, we use-DCCPP_SUITESinrt.confSimilar changes have been made to32BIT,DEBUG,REPRO,MULTI_GASES, etc.Issue(s) addressed
Part of #416
More work might be needed based on the comments to this PR.
Testing
How were these changes tested? What compilers / HPCs was it tested with? Are the changes covered by regression tests? (If not, why? Do new tests need to be added?) Have regression tests and unit tests (utests) been run? On which platforms and with which compilers? (Note that unit tests can only be run on tier-1 platforms)
Dependencies
None.