On Tue, 5 Aug 2008, Yann E. MORIN wrote:

> I'm using uClibc 20080801, and I did set UCLIBC_SUPPORT_AI_ADDRCONFIG=y
> in my uClibc config file, thus getifaddrs get compiled in uClibc.so.
>
> To check if it can use getifaddrs et al., the Java configure tries to link
> a program that calls getifaddrs, but does not include ifaddrs.h, it just
> defines a prototype on its own. Because getifaddrs does in fact exist in
> uClibc.so, the link is successfull, and configure #defines HAVE_GETIFADDRS
> to 1. The calls to getifaddrs and freeifaddrs are then guarded with
> ...
> But when Java gets built, it fails with:
> ...
> Before r22534, there was a include/ifaddrs.h file. At 22534 this file was
> moved to libc/inet/ifaddrs.h as it was supposed to be a private header:
> ...
> But that is not true in fact. As for glibc, it does install this header.
>
> ...
>
> Can we revert 22534, or is there a better way?

include/ifaddrs.h was not the same as libc/inet/ifaddrs.h ; the former 
included the latter and thus could not be included outside uclibc. The 
files were taken from glibc; rather than try and figure out the necessary 
infrastructure to get it working, we moved the contents of 
include/ifaddrs.h to libc/inet/ifaddrs.h and removed the former, since we 
didn't need it outside uclibc anyway.

However, I can't see a problem with making it globally accessible, so the 
change you propose

> So I moved ifaddrs.h back to the include/ directory, also changed back
> libc/inet/ifaddrs.c and libc/inet/getaddrinfo.c to #include <ifaddrs.h>
> (instead of "ifaddrs.h"), and the build completes.

should be ok. Check that the whole shebang builds properly if 
UCLIBC_SUPPORT_AI_ADDRCONFIG is not defined. Also, technically it's not 
reverting 22534 as the files looked different then.

Then again, someone else may have another view. Seems a bit odd of a 
configuration script to define its own prototype and try to link, rather 
than just trying to see if the requested function could be found in an 
include file, but I guess there's a good reason for doing so.

/Ricard
--
Ricard Wolf Wanderlöf                           ricardw(at)axis.com
Axis Communications AB, Lund, Sweden            www.axis.com
Phone +46 46 272 2016                           Fax +46 46 13 61 30
_______________________________________________
uClibc mailing list
uClibc@uclibc.org
http://busybox.net/cgi-bin/mailman/listinfo/uclibc

Reply via email to