[PATCH] D134168: [RISCV] Make preferred alignment of PointerArgs for MemIntrinsic

Jessica Clarke via Phabricator via llvm-commits llvm-commits at lists.llvm.org
Mon Sep 19 19:25:33 PDT 2022


jrtc27 added a comment.

> Set default preferred alignment for MemIntrinsic like memcpy according to arch32 or arch64,
> it will improve performance.

What are arch32 and arch64?

> e.g. dhrystone with "-O2" boosts performance by 50% on arch RV32.

On what implementation? Does this affect actually-useful benchmarks, not just dhrystone? I would assume so, but it'd be more useful to get numbers for meaningful benchmarks rather than ones people should've long since abandoned.



================
Comment at: llvm/lib/Target/RISCV/RISCVISelLowering.cpp:12986
+    return false;
+  MinSize = 8;
+  PrefAlign = Align(Subtarget.getXLen() / 8);
----------------
This seems arbitrary. I'd expect at least XLEN / 8 here?


CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D134168/new/

https://reviews.llvm.org/D134168



More information about the llvm-commits mailing list