[Bug c++/81327] [8 Regression] cast to void* does not suppress -Wclass-memaccess

[ville.voutilainen at gmail dot com]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=81327

--- Comment #15 from Ville Voutilainen <ville.voutilainen at gmail dot com> ---
I would like to have the change be a DR. Anyway, I looked at the cast-less
cases, they are in qtransform.h, which is included by at least QImage and
QtGui, so there will be a fair amount of code that will hit the problem. It's
likely less severe than the casting case, but quite annoying still. This will
also probably cause churn in linux distros when they update their compiler.

I'm not a fan of taking the warning out of -Wall. But I think it's probably the
less painful approach for existing code.