[PATCH] gfortran -- Map REAL128 to REAL kind type with widest precision

[Steve Kargl]

All,

The attach patch maps REAL128 from iso_fortran_env to 
the REAL kind type with widest precision.  Prior to
this patch, REAL128 is mapped to the first REAL kind
type with a matching storage size of 128 bits.  On
x86_64-*-freebsd both REAL(10) and REAL(16) occupy 
16 bytes, but REAL(10) is mapped to REAL128.  This 
may surprise users that assume REAL128 is mapped to
an IEEE754-2008 128-bit floating point entity with
113 bits of precision.  OK to commit?

2017-02-07  Steven G. Kargl  <kargl@gcc.gnu.org>

	* trans-types.c	(gfc_get_int_kind_from_width_isofortranen):  Choose
	REAL type with the widest precision if two (or more) have the same
	storage size.

-- 
Steve
20161221 https://www.youtube.com/watch?v=IbCHE-hONow
-------------- next part --------------
A non-text attachment was scrubbed...
Name: trans_types.diff
Type: text/x-diff
Size: 1604 bytes
Desc: not available
URL: <http://gcc.gnu.org/pipermail/gcc-patches/attachments/20170207/dae25515/attachment.bin>