[PATCH] D20762: AArch64: Do not test for CPUs, use SubtargetFeatures
Arnaud A. de Grandmaison via llvm-commits
llvm-commits at lists.llvm.org
Thu Jun 2 10:45:24 PDT 2016
aadg accepted this revision.
aadg added a comment.
This looks OK to me from a PBQP perspective (see comment).
================
Comment at: lib/Target/AArch64/AArch64Subtarget.cpp:167
@@ -140,6 +166,3 @@
AArch64Subtarget::getCustomPBQPConstraints() const {
- if (!isCortexA57())
- return nullptr;
-
- return llvm::make_unique<A57ChainingConstraint>();
+ return balanceFPOps() ? llvm::make_unique<A57ChainingConstraint>() : nullptr;
}
----------------
FP load balancing has to be done with a specific pass when using the GreedyRegAlloc. PBQPRegAlloc can do it right away (no need for a separate pass), but the allocator need to learn the FP load balancing constraint it has to solve (on top of the "standard" constraints).
Enabling it for both A57 and A53 is I think the right thing to do.
Repository:
rL LLVM
http://reviews.llvm.org/D20762
More information about the llvm-commits
mailing list