Fix for #6903 - enable vectorization of loops with non-constant lower bound. #6926
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
The change avoids creating multiple induction variables, which seem to thwart LLVM 3.3's vectorizer (but not LLVM 3.4).
JeffBezanson
added a commit
that referenced
this pull request
May 22, 2014
Fix for #6903 - enable vectorization of loops with non-constant lower bound.
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Outdated suggestions cannot be applied.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.
This improvements enables vectorization of the #6903. It also improves performance of some of my other
@simdbenchmarks.I timed 5 different variations of loop lowering, including the normal one that Julia uses, and this is one of three variations that came out the fastest (within measurement noise) for aforementioned benchmarks, and did significantly better than using Julia's normal loop-lowering scheme.
When we get to LLVM 3.5, let's try using Julia's usual loop lowering. LLVM 3.4's vectorizer seemed to do much better than LLVM 3.3 with the usual lowering.