-
Notifications
You must be signed in to change notification settings - Fork 54
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
remedy segfault with openmpi/intelmpi in MPI_Reduc #115
base: master
Are you sure you want to change the base?
Conversation
2.) dgemm take heed of 'rectangularism' of contraction 3.) used realistic model values for vsc4
I am a bit confused about the scope of changes in this request. Seems like mostly stuff for dry-runs, but also changes to model parameters are included (possibly due to more recent CTF commits?). Some tweaks to the time estimates seem like they would affect performance nontrivially, on the other hand if they are corrections/improvements, this might be valuable. Sorry for not getting to this for such a long time. |
Sorry I messed something up with my pull request. Please dont accept it |
- minor changes in the output of dry-run - switch on exhaustive in (almost) every contraction - dont concider 'shape' in gemm contractions - new feature: only add model parameters which were actually tuned - change: dump only the slowest measurement from all ranks
No description provided.