[PATCH] D14489: [AArch64] Applying load pair optimization for volatile load/store
Junmo Park via llvm-commits
llvm-commits at lists.llvm.org
Wed Nov 11 16:43:34 PST 2015
flyingforyou added a comment.
Hi Chad.
Could you give me an idea that how to change the code without limiting scan value to 1?
Should I make another function for merging adjacent volatile ldr/str?
Or move "hasOrderedMemoryRef" function to inside of "findMatchingInsn" and
change the Limit value 1 if the first instruction is volatile.
First one makes many duplication of codes. You may not be satisfied with second one.
If you have any other opinion, tell me please.
Thanks.
http://reviews.llvm.org/D14489
More information about the llvm-commits
mailing list