>>88020019
Yeah, that would be nice.
It's kind of ironic that all this mess is BECAUSE glibc cares about backwards compatibility so much. It keeps shipping multiple versions of many functions that behave slightly differently or contain certain bugs that software depends on. If there was no versioning there wouldn't be a problem...
Actually, it looks like you can do something like this:
__asm__(".symver realpath,realpath@GLIBC_2.2.5");
and it'll use the realpath function versioned at 2.2.5.