Skip to content

Conversation

@sjakobi
Copy link
Member

@sjakobi sjakobi commented Oct 19, 2025

(Extracted from #521)

This speeds up fromList about 1–2% and prepares the ground for #520.

This reduces the core size for the inner loop of two by 10 terms.
* Use the "shifted hash" approach.
* Create a variant `two'` that can be used to address #468.
Just to be safe.
@sjakobi sjakobi mentioned this pull request Oct 19, 2025
2 tasks
@sjakobi
Copy link
Member Author

sjakobi commented Oct 21, 2025

After some additional benchmark runs und more staring at the Core diffs, I've become less convinced that this is clear-cut improvement. While I'm fairly convinced that the inner loop of two is now faster, I suspect that this hardly matters, since the recursive case is only rarely triggered.

On the other hand, it may matter that the allocation of the new Leaf is now inlined into the calling function.

@sjakobi sjakobi marked this pull request as draft October 21, 2025 10:40
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

1 participant