[llvm] 50d38cf - [X86] X86FixupVectorConstants.cpp - update comment to describe all the constant load ops performed by the pass

Simon Pilgrim via llvm-commits llvm-commits at lists.llvm.org
Wed Feb 7 06:07:21 PST 2024


Author: Simon Pilgrim
Date: 2024-02-07T14:07:02Z
New Revision: 50d38cf934861205768de9e3cf32827e73764001

URL: https://github.com/llvm/llvm-project/commit/50d38cf934861205768de9e3cf32827e73764001
DIFF: https://github.com/llvm/llvm-project/commit/50d38cf934861205768de9e3cf32827e73764001.diff

LOG: [X86] X86FixupVectorConstants.cpp - update comment to describe all the constant load ops performed by the pass

Added: 
    

Modified: 
    llvm/lib/Target/X86/X86FixupVectorConstants.cpp

Removed: 
    


################################################################################
diff  --git a/llvm/lib/Target/X86/X86FixupVectorConstants.cpp b/llvm/lib/Target/X86/X86FixupVectorConstants.cpp
index f65fa5a2298ac8..9c46cee572fc91 100644
--- a/llvm/lib/Target/X86/X86FixupVectorConstants.cpp
+++ b/llvm/lib/Target/X86/X86FixupVectorConstants.cpp
@@ -8,9 +8,10 @@
 //
 // This file examines all full size vector constant pool loads and attempts to
 // replace them with smaller constant pool entries, including:
-// * Converting AVX512 memory-fold instructions to their broadcast-fold form
+// * Converting AVX512 memory-fold instructions to their broadcast-fold form.
+// * Using vzload scalar loads.
 // * Broadcasting of full width loads.
-// * TODO: Zero extension of full width loads.
+// * Sign/Zero extension of full width loads.
 //
 //===----------------------------------------------------------------------===//
 


        


More information about the llvm-commits mailing list