-
Notifications
You must be signed in to change notification settings - Fork 17
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
fixes #2657 #2658
fixes #2657 #2658
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.
@bena-nasa I think this needs to be changed to nonzero diff. If I understood correctly it really can change answers.
Also ... does this possibly impact the results that the modeling team reported yesterday? Were we shaving one of their fields or only a diagnostic?
Converting to draft. Per @biljanaorescanin the debug runs of the LDAS are throwing a failure: GEOS-ESM/GEOSldas#734 (comment) Intel:
GNU:
I suppose the "good" thing is it's the same error it seems. |
@tclune I don't think anyone in the AGCM world operates in such a way to trigger this. But I'll work with @bena-nasa to get specifics on when this could be triggered so we can let people know in the release notes and changelog. |
@biljanaorescanin |
need to copy whole state
The CI failure looks to be some weird issue on their side. I'm re-launching |
With last commit all my land tests passed. |
I did some spot checking of the output files by using that converter, made sure that the shaved vs non-shaved output was right in that each variable was "right" in that they were basically the same modulo the bit shaving. I.E. I didn't do something like somehow write the variables in the wrong order in the binary file in say the shaved path. |
Description
Related Issue
As the title says this fixes #2657
If the the binary output is bit shaved, I copy to a new state before writing.
In reality this is only necessary for instantaneous output but I was trying to avoid a whole slew of conditions and it doesn't matter in the end if we make a few extra copies
I've marked this as zero-diff because is is zero-diff for the GEOSgcm.x although it is not for the LDAS.
Motivation and Context
How Has This Been Tested?
Types of changes
Checklist: