[llvm-dev] Forcing llvm to not wrap two 32bit accesses into a single 64bit access

[Tim Northover via llvm-dev]

On Wed, 19 May 2021 at 13:51, James Courtier-Dutton via llvm-dev
<llvm-dev at lists.llvm.org> wrote:
> How do I ensure that the optimisation will only optimise to 32bit wide
> instructions?

The only way I can think of is by marking them volatile (and possibly
hand-promoting to 32-bit if that matters to you). Or assembly of one
form or another of course.

Cheers.

Tim.