[PATCH] D12681: Calling conventions for HHVM

Maksim Panchenko via llvm-commits llvm-commits at lists.llvm.org
Mon Sep 28 17:02:43 PDT 2015


maksfb added a comment.

Thanks for the review, Quentin!


================
Comment at: include/llvm/Support/MathExtras.h:621
@@ +620,3 @@
+                                   uint64_t Skew = 0) {
+  Skew %= Align;
+  return (Value + Align - 1 - Skew) / Align * Align + Skew;
----------------
qcolombet wrote:
> In the comment you said that Skew must be less than Align.
> So, this line of code seems useless.
> Either keep that line and change the comment to actually reflect what is happening or change that into an assert Skew < Align.
Good point. I'll change the comment.

================
Comment at: lib/CodeGen/PrologEpilogInserter.cpp:574
@@ +573,3 @@
+  unsigned Skew = 0;
+  if (LLVM_UNLIKELY(Fn.getFunction()->getCallingConv() == CallingConv::HHVM))
+    Skew = 8;
----------------
qcolombet wrote:
> I would rather have a target hook for that: getSkew, so that we do not expose (more) target code in the generic part.
Makes sense. Will move the logic to `TargetFrameLowering`.


http://reviews.llvm.org/D12681





More information about the llvm-commits mailing list