[Ada] Improper behavior of floating-point attributes

[Ada] Improper behavior of floating-point attributes

This patch fixes an error in the handling of attributes Pred qnd Succ when
applied to the limit values of a floating-point type. The RM mandates that
such operations must raise constraint_error, but GNAT generated in most cases
an infinite value, regardless of whether overflow checks were enabled.

2018-05-29  Ed Schonberg  <schonberg@adacore.com>

gcc/ada/

	* libgnat/s-fatgen.adb (Succ, Pred):  Raise Constraint_Error
	unconditionally when applied to the largest positive (resp. largest
	negative) value of a floating-point type.

gcc/testsuite/

	* gnat.dg/float_attributes_overflows.adb: New testcase.

From-SVN: r260882