[LLVMdev] [RFC] Add compiler scheduling barriers

Yi Kong kongy.dev at gmail.com
Thu Jul 3 04:04:20 PDT 2014


After some internal discussions, we concluded that moving
non-memory-access and non-side-effect instructions around memory
barrier intrinsics is fine in almost any typical situation. In the
rare occasions where reordering is undesirable, the programmer should
use inline asm instead.

Thus I'm dropping this RFC. Thanks for your comments.

-Yi



More information about the llvm-dev mailing list