cmd/compile: high inline cost of encoding/binary and math.Float32bits #42788
Comments
As far as the Float32bits part goes, it could be similar to #42739. |
The big difference in BTW, |
Thanks, I'm aware. The code is a simplification of #28941 (comment) to ensure BCE works. |
I was thinking about the conversion overhead in math.Float32bits. Maybe I currently added a new case in
I'm not entirely sure, whether the
It didn't make it equivalent to just returning, however it seems like a worthwhile improvement nevertheless. |
I guess the whole This is more hacky case than the previous:
|
Perhaps all |
Yeah, that is simpler. I wasn't sure whether string to byte (or any other similar) conversions are represented by |
Change https://golang.org/cl/281232 mentions this issue: |
What version of Go are you using (
go version
)?Does this issue reproduce with the latest release?
Yes, also happens on tip.
What did you do?
The inline cost calculation for these seem inconsistent https://go.godbolt.org/z/YMKWjc
The inline cost for those:
There seems to be a really large difference between EncodeQuad2, EncodedQuad3 and EncodeQuad. I would expect all of them to be inlinable and similar in cost.
The text was updated successfully, but these errors were encountered: