[PATCH] D54355: Use is.constant intrinsic for __builtin_constant_p

Mon Dec 10 21:09:48 PST 2018

void added a comment.

In D54355#1326494 <https://reviews.llvm.org/D54355#1326494>, @craig.topper wrote:

> I'm seeing a case where an inline assembly 'n' constraint is behaving differently in -O0 now. Previously we would evaluate the __builtin_constant_p as part of the EvaluateAsInt call in CodeGenFunction::EmitAsmInput, but I think we're not doing that now. This causes us to fail later because the constraint wasn't satisfied since we were in -O0 and thus we didn't optimize the rest of the expression. The test case we have is a nasty mix of __builtin_constant_p, __builtin_choose_expr, and __builtin_classify_type. I can see about sharing it if needed.

`__builtin_constant_p()` is known to have different semantics at `-O0` than it does during optimizations. That said, we shouldn't be intentionally generating incorrect code. Could you test it against gcc to see if it has the same semantics? Otherwise, please add a testcase. :-)

Repository:
  rC Clang

CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D54355/new/

https://reviews.llvm.org/D54355