[llvm] r242714 - [ARM] Refactor the prologue/epilogue emission to be more robust.

[Tim Northover via llvm-commits]

On 30 November 2015 at 09:42, Quentin Colombet via llvm-commits
<llvm-commits at lists.llvm.org> wrote:
> Maybe the test or check line is just poorly written because the code that does this optimization is guarded under  STI.hasV5TOps(), i.e., this should not happen for v4t only arch.

They both look fine to me: the triple is armv4t when "CHECK-V4T" is
used, and the code sequence does have the problems Jon mentions.

Cheers.

Tim.