[Bug middle-end/84095] [8 Regression] false-positive -Wrestrict warnings for memcpy within array

[msebor at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=84095

--- Comment #16 from Martin Sebor <msebor at gcc dot gnu.org> ---
Thanks for the test case in comment #14.  I've reproduced the warning and will
look into it.

The false positive from comment #9 isn't fixed yet.  I'm assuming it's due to
the same root cause as bug 83456 that I'm tracking separately.  There's no way
to avoid that one other than by disabling the warning in a whole set of cases. 
I'm trying to decide if trading the false positives for the unavoidable false
negatives make sense.