[Bug c/31924] New: gcc accepts invalid suffixes for decimal float constants

[janis at gcc dot gnu dot org]

The change to the C preprocessor to recognize "df", "dd", and "dl" as suffixes
for decimal floating point constants also allows "fd" and "ld"; a check for
that assumed that the the letters are examined from left to right, but they are
examined from right to left.  There are no errors for the following:

  double a = 1.ld;
  double b = 1.fd;

I'm testing a fix, but wanted a PR to reference for a patch to the 4.2 branch.


-- 
           Summary: gcc accepts invalid suffixes for decimal float constants
           Product: gcc
           Version: 4.2.0
            Status: UNCONFIRMED
          Keywords: accepts-invalid
          Severity: normal
          Priority: P3
         Component: c
        AssignedTo: janis at gcc dot gnu dot org
        ReportedBy: janis at gcc dot gnu dot org


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=31924