[Bug c/84717] New: C compiler accepts fortran style suffix "d" with float number but result is strange

[stiefschmied at alice dot de]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=84717

            Bug ID: 84717
           Summary: C compiler accepts fortran style suffix "d" with float
                    number but result is strange
           Product: gcc
           Version: 6.4.0
            Status: UNCONFIRMED
          Severity: normal
          Priority: P3
         Component: c
          Assignee: unassigned at gcc dot gnu.org
          Reporter: stiefschmied at alice dot de
  Target Milestone: ---
            Target: x86_64-apple-darwin (and others?)

Created attachment 43564
  --> https://gcc.gnu.org/bugzilla/attachment.cgi?id=43564&action=edit
example c code which compiles with gcc 6.4.0 but should give an error.

When porting code from Fortran, a floating point number was accidentially
written like this:
   static double aem = 7.29735d-3;

gcc version 6.4.0 accepts the code, but assigns an unexpected result:
   static double aem = 7.29735-3;

Other compilers (tested with: clang 9.0.0 or gcc version 4.4.7) report an error
during compilation:

error: invalid suffix "d" on floating constant

I think this is a compiler bug? The example code is attached.