CodeGen: Use more efficient lowering for UNM_* #1177
Closed
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
UNM_NUM and UNM_VEC were both implemented assuming SSE-style restrictions (2-argument form), but using AVX that doesn't have them. There's no need to copy source to destination separately - we can just vxorpd into destination. Most occurrences of UNM_NUM/UNM_VEC followed the self-xor path, but this saves a couple instructions in trig benchmark and makes it execute ~0.1% fewer instructions (the actual runtime delta is within the noise).
vegorov-rbx
reviewed
Mar 4, 2024
| build.vxorpd(inst.regX64, inst.regX64, build.f64(-0.0)); | ||
| } | ||
|
|
||
| build.vxorpd(inst.regX64, regOp(inst.a), build.f64(-0.0)); |
There was a problem hiding this comment.
Please flag the changes, as we are raising code flagging requirements internally.
There was a problem hiding this comment.
Closing this, feel free to make this change internally.
There was a problem hiding this comment.
We have added these changes internally.
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
UNM_NUM and UNM_VEC were both implemented assuming SSE-style restrictions (2-argument form), but using AVX that doesn't have them. There's no need to copy source to destination separately - we can just vxorpd into destination.
Most occurrences of UNM_NUM/UNM_VEC followed the self-xor path, but this saves a couple instructions in trig benchmark and makes it execute ~0.1% fewer instructions (the actual runtime delta is within the noise).