[Bug regression/63177] Powerpc no-vfa-vect-depend-2.c and no-vfa-vect-depend-3.c failures

[rguenth at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=63177

--- Comment #3 from Richard Biener <rguenth at gcc dot gnu.org> ---
The version-for-alias code makes choices dependent on ordering of data
references
which changes from time to time.  I've repeatedly tried to clean up the code
but only manage to break more testcases where the things they are testing are
often odd (and wrong?).