re PR c/34252 (DEC32_MIN_EXP, DEC32_MAX_EXP don't match TR 24732)

gcc/
	PR c/34252
	* ginclude/float.h: Rename DECnn_DEN to DECnn_SUBNORMAL_MIN.
	* real.c (decimal_single_format): Correct values of emin and emax.
	(decimal_double_format): Ditto.
	(decimal_quad_format): Ditto.
	* c-cppbuiltin.c (builtin_define_decimal_float_constants): Adjust
	computation of DECnn_MIN and DECnn_MAX for corrected values of
	emin and emax.  Define __DECnn_SUBNORMAL_MIN__ instead of
	__DECnn_MIN__, and adjust its computation for the corrected value
	of emin.
	
gcc/testsuite/
	PR c/34252
	* gcc.dg/dfp/decfloat-constants.c: Check for DECnn_SUBNORMAL_MIN
	instead of DECnn_DEN.  Support -DDBG to list lines that fail.

From-SVN: r143128