Closed jedbrown closed 4 years ago
@JustinTArthur It looks like you approved a very similar set of changes before in: https://github.com/RadeonOpenCompute/ROC-smi/pull/80
Do you know if there a reason we didn't incorporate those changes?
CC @kentrussell ?
We ended up missing the 3.5 cutoff due to some critical bugs. I'm trying to get the fix for this issue (IE my poor choice in operators) into 3.6 Basically everything is high priority, and since we didn't have Python3.8 on the test machines in automation, it got bumped down in priority. Either way I've brought it internally and it's undergoing testing. I'll close off both of these PRs once it's released publicly. Sorry for the delay
Fixed in 3.6
is
shouldn't be used in this context due to surprising results such asCc: @moes1