[avahi] cool debug trap
Padraig O'Briain
Padraig.Obriain at Sun.COM
Tue Aug 22 06:14:44 PDT 2006
Lennart Poettering wrote:
> On Tue, 22.08.06 09:50, Padraig O'Briain (Padraig.Obriain at Sun.COM) wrote:
>
>
>> The Makeifile.am files contain a cool debug trap for i396/gcc.
>>
>> I needed to remove it when building on Solaris.
>>
>
> Why that? It's just a macro. As long as nobody uses it it shouldn't
> influence compilation in any way.
>
> Could you please paste the exact error text that is shown when
> compiling it on Solaris?
>
>
I get the problem when doing make install, for example in avahi-core
directory. The result is that rebuild to not refer to libraries in the
source tree does not occur.
bash-3.00$ make install
/bin/sh ../libtool --tag=CC --mode=link
/sgnome/tools/x86-solaris/forte/venus-20.2/opt/SUNWspro/bin/cc -I..
'-DDEBUG_TRAP=__asm__("int $3")'
-DAVAHI_DAEMON_RUNTIME_DIR=\"/var/run/avahi-daemon/\"
-DAVAHI_SOCKET=\"/var/run/avahi-daemon/socket\"
-DAVAHI_SERVICE_DIR=\"/etc/avahi/services\"
-DAVAHI_CONFIG_FILE=\"/etc/avahi/avahi-daemon.conf\"
-DAVAHI_HOSTS_FILE=\"/etc/avahi/hosts\"
-DAVAHI_DBUS_INTROSPECTION_DIR=\"/usr/share/avahi/introspection\"
-DAVAHI_CONFIG_DIR=\"/etc/avahi\" -i -xO4 -xspace -xstrconst -xpentium
-mr -xregs=no%frameptr -D_XOPEN_SOURCE=500 -D__EXTENSIONS__
-I/usr/sfw/include -Wl,-zignore -Wl,-zcombreloc -Wl,-Bdirect
-L/usr/lib/mdns -R/usr/lib/mdns -ldns_sd -lsocket -lnsl -L/usr/sfw/lib
-R/usr/sfw/lib -lexpat -o ini-file-parser-test
ini_file_parser_test-ini-file-parser.o
ini_file_parser_test-ini-file-parser-test.o
../avahi-common/libavahi-common.la ../avahi-core/libavahi-core.la
../libtool: \\)" -i -xO4 -xspace -xstrconst -xpentium -mr
-xregs=no%frameptr -D_XOPEN_SOURCE=500 -D__EXTENSIONS__
-I/usr/sfw/include -Wl,-zignore -Wl,-zcombreloc -Wl,-Bdirect
-L/usr/lib/mdns -R/usr/lib/mdns -ldns_sd -lsocket -lnsl -L/usr/sfw/lib
-R/usr/sfw/lib -lexpat -o libavahi-core.la -rpath /usr/lib
-export-dynamic -version-info 4:3:0 libavahi_core_la-timeeventq.lo
libavahi_core_la-iface.lo libavahi_core_la-server.lo
libavahi_core_la-entry.lo libavahi_core_la-prioq.lo
libavahi_core_la-cache.lo libavahi_core_la-socket.lo
libavahi_core_la-response-sched.lo libavahi_core_la-query-sched.lo
libavahi_core_la-probe-sched.lo libavahi_core_la-announce.lo
libavahi_core_la-browse.lo libavahi_core_la-rrlist.lo
libavahi_core_la-resolve-host-name.lo
libavahi_core_la-resolve-address.lo libavahi_core_la-browse-domain.lo
libavahi_core_la-browse-service-type.lo
libavahi_core_la-browse-service.lo libavahi_core_la-resolve-service.lo
libavahi_core_la-dns.lo libavahi_core_la-rr.lo libavahi_core_la-log.lo
libavahi_core_la-browse-dns-server.lo libavahi_core_la-fdutil.lo
libavahi_core_la-util.lo libavahi_core_la-hashmap.lo
libavahi_core_la-wide-area.lo libavahi_core_la-multicast-lookup.lo
libavahi_core_la-querier.lo libavahi_core_la-addr-util.lo
libavahi_core_la-domain-util.lo libavahi_core_la-iface-pfroute.lo
../avahi-common/libavahi-common.la @inst_prefix_dir@): cannot execute
>> Is there a way to patch the filess so that those who want the debug trap
>> get it and those who don't want it don't get it?
>>
>
> Hmm. I don't think this is necessary. A call to the debug trap should
> never be commited to SVN. It's only there for debugging, i.e. sticking
> it into some code and testing and after everything works removing it
> again.
>
> Lennart
>
>
More information about the avahi
mailing list