[
https://issues.apache.org/jira/browse/TS-1768?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13623879#comment-13623879
]
James Peach commented on TS-1768:
---------------------------------
Wow, this turned into a big mess.
It looks like we actually link -lresolv by fluke. At least on Darwin, libc
contains the resolver functions we check for, but not the ones we use. We need
extra m4 to check for these because Bind 9 #defines the API names to
version-specific symbols (eg, dn_comp becomes res_9_dn_comp).
There also may be some issues with AC_SEARCH_LIBS always updating LIBS. We
typically put the results of AC_CHECK_LIB in a special variable, so we might
need a TS_SEARCH_LIBS to do this if the difference ends up mattering.
> use AC_SEARCH_LIBS instead of AC_CHECK_LIB
> ------------------------------------------
>
> Key: TS-1768
> URL: https://issues.apache.org/jira/browse/TS-1768
> Project: Traffic Server
> Issue Type: Improvement
> Components: Build
> Reporter: James Peach
> Assignee: James Peach
> Fix For: 3.3.4
>
>
> In general, we ought to be using AC_SEARCH_LIBS, for the reasons outlined
> here: http://www.flameeyes.eu/autotools-mythbuster/autoconf/finding.html. We
> will need to check each of the 40 or so AC_CHECK_LIB tests just to make sure
> the replacement is safe.
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators
For more information on JIRA, see: http://www.atlassian.com/software/jira
