[Bug rtl-optimization/92342] [10 Regression] a small missed transformation into x?b:0

[law at redhat dot com]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=92342

--- Comment #14 from Jeffrey A. Law <law at redhat dot com> ---
That wouldn't be a big surprise Andrew.

My point is I think we could create some match.pd patterns to canonicalize the
form in gimple and it wouldn't matter nearly as much what combine did or didn't
do to this kind of code.