You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
{{ message }}
This repository has been archived by the owner on Jun 28, 2022. It is now read-only.
I'm in the process of implementing a very nice an "obvious" simplification: using the same number and arrangement/mapping of virtual registers to spill slots across all fragment partitions in a trace. This opens up the obvious question/hope/idea: now that the structure of spill slots is identical across fragment partitions: can we merge all the spilling/filling across partitions, where there once was no such merging? This would be a super nice things to have :-D
The text was updated successfully, but these errors were encountered:
…tions to stack analysis: if *any* fragment is invalid, then all are. Otherwise, assume all are valid. This is such a great simplification, I wish I had thought of it sooner!
Sign up for freeto subscribe to this conversation on GitHub.
Already have an account?
Sign in.
I'm in the process of implementing a very nice an "obvious" simplification: using the same number and arrangement/mapping of virtual registers to spill slots across all fragment partitions in a trace. This opens up the obvious question/hope/idea: now that the structure of spill slots is identical across fragment partitions: can we merge all the spilling/filling across partitions, where there once was no such merging? This would be a super nice things to have :-D
The text was updated successfully, but these errors were encountered: