Revert multiply to use non-intrinsic exactness checks #8523
Conversation
This file contains hidden or 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
Member
Author
|
Note that JRuby 10 could use a different patch proposed in #8516 that uses |
Member
Author
|
Ok clearly I botched this. Will circle back. |
The call to Math.multiplyExact works very well when the result does not overflow, skipping range checks and using the host CPU's support for overflow checks. However when we are dancing on the range boundary and multiplyExact fails repeatedly, the cost of raising all those exceptions massively outweighs the gains. This is the primary cause of performance issues reported in jruby#8516. This patch reverts to a version of the old range-checking logic. Performance on the bug report's benchmark is an order of magnitude faster for the two cases that seemed to overflow a lot, and the other cases do not seem to degrade. Fixes jruby#8516
663b0ac to
b6f3b05
Compare
Member
Author
|
New patch looks clean enough. I'm only doing this one targeted fix since we have alternatives in JRuby 10 using Java 21. |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
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Outdated suggestions cannot be applied.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.
The call to Math.multiplyExact works very well when the result does not overflow, skipping range checks and using the host CPU's support for overflow checks. However when we are dancing on the range boundary and multiplyExact fails repeatedly, the cost of raising all those exceptions massively outweighs the gains.
This is the primary cause of performance issues reported in #8516.
This patch reverts to a version of the old range-checking logic. Performance on the bug report's benchmark is an order of magnitude faster for the two cases that seemed to overflow a lot, and the other cases do not seem to degrade.
Fixes #8516