[PATCH] [SLPVectorization] Enhance Ability to Vectorize Horizontal Reductions from Consecutive Loads

Michael Zolotukhin mzolotukhin at apple.com
Fri Dec 19 15:54:32 PST 2014


REPOSITORY
  rL LLVM

================
Comment at: lib/Transforms/Vectorize/SLPVectorizer.cpp:444
@@ +443,3 @@
+  /// while Right had v1 and v3, which prevented bundling them into
+  /// a vector of loads. Rorder them so that Left now has v0 and v1
+  /// while Right has v2 and v3 enabling their bundling into a vector.
----------------
s/Rorder/Reorder/

http://reviews.llvm.org/D6675

EMAIL PREFERENCES
  http://reviews.llvm.org/settings/panel/emailpreferences/






More information about the llvm-commits mailing list