Builderrors worked around by removing -Wshadow (error: declaration of ‘index’ shadows a global declaration) and adding -D_BSD_SOURCE (s6_addr32 not working otherwise) to AM_CFLAGS.
It seems that the IN6_ADDR_* macro are defined with s6_addr32 if __GNUC__ is used (which is always the case for gcc) and the s6_addr32 in in6_addr structure are only defined if __USE_MISC or __USE_GNU.
So it seems eglic has introduced this "bug" recently. Can you report that to them or Ubuntu team ?
TurnServer respects the ISO C99 and POSIX standard and I don't want to use GNU or other non-standard extension (__BSD_SOURCE), so I will try to see if something can be done to override this in the TurnServer code.
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
"
As the API matured and stabilized, it was incorporated into the Open
Group's Networking Services (XNS) specification, issue 5.2, which was
subsequently incorporated into a joint Open Group/IEEE/ISO standard
[3].
...
...
[3] IEEE Std. 1003.1-2001 Standard for Information Technology --
Portable Operating System Interface (POSIX). Open Group
Technical Standard: Base Specifications, Issue 6, December 2001.
ISO/IEC 9945:2002. http://www.opengroup.org/austin
"
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
I finally decided to modify source code to make it possible to compile on recent glibc (now 2.17 on Debian testing), by adding the _SVID_SOURCE compilation flag.
Fix is in SVN revision 235.
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
Can you post full compilation logs ?
Logs uploaded
It is strange because the code don't use s6_addr32. Can you post your /usr/include/netinet/in.h, please ?
You find it at: svn cat svn://svn.eglibc.org/trunk/libc/inet/netinet/in.h
#ifdef __GNUC__
# define IN6_IS_ADDR_UNSPECIFIED(a) \
(__extension__ \
({ const struct in6_addr *__a = (const struct in6_addr *) (a); \
__a->s6_addr32[0] == 0 \
&& __a->s6_addr32[1] == 0 \
&& __a->s6_addr32[2] == 0 \
&& __a->s6_addr32[3] == 0; }))
...
OK, I understand now.
It seems that the IN6_ADDR_* macro are defined with s6_addr32 if __GNUC__ is used (which is always the case for gcc) and the s6_addr32 in in6_addr structure are only defined if __USE_MISC or __USE_GNU.
So it seems eglic has introduced this "bug" recently. Can you report that to them or Ubuntu team ?
TurnServer respects the ISO C99 and POSIX standard and I don't want to use GNU or other non-standard extension (__BSD_SOURCE), so I will try to see if something can be done to override this in the TurnServer code.
Do you have a reference to the part of the ISO C99 or POSIX standard defining the IN6_ADDR_* macros in that case?
It is defined in the RFC 3493
"
As the API matured and stabilized, it was incorporated into the Open
Group's Networking Services (XNS) specification, issue 5.2, which was
subsequently incorporated into a joint Open Group/IEEE/ISO standard
[3].
...
...
[3] IEEE Std. 1003.1-2001 Standard for Information Technology --
Portable Operating System Interface (POSIX). Open Group
Technical Standard: Base Specifications, Issue 6, December 2001.
ISO/IEC 9945:2002. http://www.opengroup.org/austin
"
Reported as Ubuntu bug #1086787: https://bugs.launchpad.net/bugs/1086787
I have report the problem on eglibc mailing-list with a patch: http://www.eglibc.org/archives/patches/msg01251.html.
I finally decided to modify source code to make it possible to compile on recent glibc (now 2.17 on Debian testing), by adding the _SVID_SOURCE compilation flag.
Fix is in SVN revision 235.