Segfaulting test-case with -flto

[Martin Liška]

On 04/19/2018 02:56 PM, Stephan Bergmann wrote:
> Not sure how -Wclass-memaccess is relevant for LTO? 

It's much easier to end up with an UBSAN that eventually causes a SEGFAULT.
That's my experience learned from Firefox.

 Anyway, had cleaned up LO master the other day to compile with upcoming GCC 8, committing various improvements (see `git log --grep=-Werror=class-memaccess`). Only a couple places seemed not so easy to fix (without introducing potential performance regressions), so they remained in the state of local TODO patches for now:

I see. As mentioned one either needs to be lucky, or to enable -flifetime-dse=1.

Thanks for following the Warnings and doing the clean-up.

Martin