-
Notifications
You must be signed in to change notification settings - Fork 11
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
Fix issue #265 #266
Fix issue #265 #266
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.
All good 👍
Just to clarify, does this mean that SOMD
is now only compatible with openmm versions 8.2+?
No, this works for any version. The previous logic was unnecessary, but still (thankfully) gives the same result. |
Despite the CI passing for the PR, it's now repeatedly failing for Windows for the devel build. I'll check next week, but presumably some dependency in the environment has changed. |
This is the exception raise during compilation of the wrappers on Windows, which worked on the CI only a few days ago:
@chryswoods: Any ideas? The This is the setup for the failing runner:
The last one that worked was:
The only difference appears to be the Runner image version, which is |
The LTCG bit is a red herring, since the same message appears in the working build. |
This PR closes #265 by changing the way that periodic box vectors are updated in the SOMD OpenMM context. This change is needed due to this issue from OpenMM. We now only need to update the default box vectors in the system, then reinitialise the context, after which the updated vectors will be propagated. Without this, calling
setPeriodicBoxVectors()
directly on the context will trigger an exception when using certain platforms, e.g. CUDA.I've added tests for this update in BioSimSpace here. Note that these will always pass in the CI since the CPU platform is unaffected by this issue. As such, the test will need to be run manually on a GPU enabled device in order to validate. (The CUDA platform is conditionally used in the tests when
CUDA_VISIBLE_DEVICES
is set.)devel
into this branch before issuing this pull request (e.g. by runninggit pull origin devel
): [y]