Fixing race condition in cell permute 2 #847
Conversation
…P : performance optimization]
…P : performance optimization]
…P : performance optimization]
olupton
left a comment
olupton
left a comment
There was a problem hiding this comment.
LGTM...I guess you have checked locally the correctness and performance with one/both of 22.5/22.7 (I don't think it's straightforward to test that in the automatic CI)?
I'm also wondering whether this change makes it easy to address
and improve the coverage a little there? What do you think?
Yes it is tested with both 22.5 & 22.7. Indeed we can include the cell permute 2 CPU version to the test. I will do it. |
…t extention / Documentation
pramodk
left a comment
pramodk
left a comment
There was a problem hiding this comment.
LGTM.
(discussed some of the questions with Christos offline)
…rain/CoreNeuron#847) * Fixing race condition in cell permute 2 : OpenACC [performance optimization] * Update Unit Test / Documentation CoreNEURON Repo SHA: BlueBrain/CoreNeuron@4a25a6b
No description provided.