On Feb 4, 2011, at 1:42 PM, William A. Rowe Jr. wrote: > On 2/4/2011 7:20 AM, Jim Jagielski wrote: >> >> On Feb 3, 2011, at 6:03 PM, William A. Rowe Jr. wrote: >> >>> On 2/3/2011 5:01 PM, Stefan Fritsch wrote: >>>> On Thursday 03 February 2011, Jim Jagielski wrote: >>>>> Don't we also need to now start linking against lresolv?? >>>>> >>>>> ..../dev/httpd-git-trunk/srclib/apr-util/libaprutil-1.la -lexpat >>>>> -liconv >>>>> /Users/jim/src/asf/code/dev/httpd-git-trunk/srclib/apr/libapr-1.la >>>>> -lpthread Undefined symbols: >>>>> "_res_9_init", referenced from: >>>>> _main in libmain.a(main.o) >>>>> ld: symbol(s) not found >>>>> collect2: ld returned 1 exit status >>>>> make[1]: *** [httpd] Error 1 >>>>> make: *** [install-recursive] Error 1 >>>>> >>>>> after this patch, build now breaks (at least os OS X). >>>> >>>> I have read the AC_SEARCH_LIBS docs [1] to mean that it adds -lresolv >>>> automatically to LIBS if needed. On my system, the test result is >>>> "none needed", so I can't really test this. >>> >>> It's probably not needed because that platform picks up the global >>> namespace, >>> and resolv was already snagged by loading libapr-1.so from httpd. >> >> Was it? From what I can see, only httpd uses res_init which in >> many systems is *just* in lresolv... > > Did Stefan's patch help? We throw away LIBS/LDFLAGS etc fairly often, and > they may need to be explicitly added. >
Haven't tried yet...
