PR bfd/14207 changes vs *-*-nacl* targets

[Alan Modra]

My guess is that you'll find the problem is due to SIZEOF_HEADERS
padding inserted by the nacl linker scripts at the start of rodata
sections.  That sort of gap is a new thing for the relro segment
calculations.  The pr14207 changes didn't really change anything for
your target, except of course the addition of the abort.

-- 
Alan Modra
Australia Development Lab, IBM