-
Notifications
You must be signed in to change notification settings - Fork 202
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
Improved --try-toolchain
overly picky
#2592
Comments
Hmm, so I'm on the fence with this one. I think that if you really want to do that you should add a The issue here for me would be that you should be only doing that if you know what you are doing, so protecting against this is reasonable default behaviour. |
+1 on @ocaisa view on this, you should take care when mapping from a full toolchain to a lower subtoolchain that has less capabilities. So, there should be a hint in the error message that mentions that you can use The error message could also be improved a bit, e.g. |
+1 sure I don't mind force. Issue is here that there quite a few easyconfigs compiled with foss or intel that don't use MKL/OpenBLAS or MPI or both; as we try to compile things minimally we often lower the toolchain. |
If you need a workaround for now you can use
since when |
I think we will soon be at a place where we can be stricter on the toolchain used for easyconfigs and put them at the right level from the outset. We would just need to support mapping up to the top level, and that should be do-able fairly easily now. |
fixed in #2605 |
The improved
--try-toolchain
now rejects cases that used to work perfectly, where the toolchain can be "downgraded" to a subtoolchain.Example:
@ocaisa
The text was updated successfully, but these errors were encountered: