[Bug c/22052] [4.0/4.1 Regression] redefinition of inline function succeeds

["jsm28 at gcc dot gnu dot org" <gcc-bugzilla at gcc dot gnu dot org>]

------- Additional Comments From jsm28 at gcc dot gnu dot org  2005-06-13 23:18 -------
There is a case the multiple static inline definitions of the same function in a
translation unit violate the constraint of  6.9#3 (no more than one external
definition of an identifier with internal linkage in the same translation unit),
if you take the definition of "inline definition" in 6.7.4#6 as only applying to
functions with external linkage.

In general C99 presumes there is no more than one definition of a function in a
translation unit (in 6.7.4#6) but has no constraint requiring this to be
diagnosed.  I think it would be reasonable in C99 mode, as part of an
implementation of C99 inline semantics, to prohibit all such cases.  inline in
gnu89 mode should remain compatible with traditional GCC semantics, which in the
case of the present bug means rejecting the problem code but in some cases may
mean accepting it (in particular where one definition is extern inline).


-- 
           What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |jsm28 at gcc dot gnu dot org
OtherBugsDependingO|                            |16989
              nThis|                            |


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=22052