Hi, >>>>> On Wed, 19 Jul 2006 18:12:15 -0600 (MDT) >>>>> "M. Warner Losh" <[EMAIL PROTECTED]> said:
imp> In message: <[EMAIL PROTECTED]> imp> Guy Helmer <[EMAIL PROTECTED]> writes: imp> : We just tried running programs under RELENG_6_1 that were compiled under imp> : RELENG_6 checked out 2006-07-19, and couldn't because of the undefined imp> : symbol "__res_state"l, which I would assume is a result of the recent imp> : MFC of the BIND 9 resolver library. Is this to be expected? It will imp> : cause a bit of a hassle... imp> It is not officially supported by the project. You are running a imp> binary compiled on a newer version of the system on an older version imp> of the system. This has sometimes worked in the past, but is outside imp> the area that's expected to work. There's been a number of breakages imp> similar to this in past RELENG branches (there was one in 3.x and a imp> lot on 4.x). Yes, I believe running RELENG_6 binary on RELENG_6_1 is not supported by the project. imp> Having said that, can someone track down the problem in more detail to imp> see if there might not be something we can do to mitigate the problem imp> in the older versions? What is __res_state? __res_state is a function to initialize actual _res variable. The users usually use _res. _res is substituted with __res_state by the macro defined in resolv.h to support multi thread application. Though it used to be ___res, it is replaced by __res_state during resolver update. We have compatibility cruft for ___res to keep ABI backward compatibility. Sincerely, -- Hajimu UMEMOTO @ Internet Mutual Aid Society Yokohama, Japan [EMAIL PROTECTED] [EMAIL PROTECTED],jp.}FreeBSD.org http://www.imasy.org/~ume/ _______________________________________________ [email protected] mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-stable To unsubscribe, send any mail to "[EMAIL PROTECTED]"
