We have not checked for this KEYWORD for a long time now, so this
is a complete noop, and thus no PORTREVISION bump. Removing it at
this point is mostly for pedantic reasons, and partly to avoid
perpetuating this anachronism by copy and paste to future scripts.
devel/distcc 2.16 -> 2.18.3
http://distcc.samba.org/ftp/distcc/distcc-2.18.3.NEWS
Consists mostly of bug fixes. Only one change in the way distcc
works. You must specify a network on which connections are
allowed, connections from all networks will still be admitted
but connections outside the specified network will be logged
and dropped.
The 127.0.0.0/8 network is the default for the port. All
existing users and all new users will need to update/add a
distccd_flags variable in /etc/rc.conf, a template for which
can be found in the rc.d script.
Perhaps a new variable, which simply contained the trusted
network address would be better, like
distccd_network="192.168.10.2/24"
PR: ports/75718
Submitted by: Dominic Marks <dom@helenmarks.co.uk>
the gnome monitor, the install failed due to remaining auto detection
stuff.
Reported by: Andrew Heybey <ath@niksun.com>
Submitted by: Clement Laforet <sheepkiller@cultdeadsheep.org>
I don't have the time to give to many of the ports I maintain at
the moment. I wasn't expecting to be so busy this summer ...
Please set all my ports back to ports@freebsd.org.
PR: ports/56935
Submitted by: Dominic Marks <dom@wirespeed.org.uk>,Clement Laforet <sheepkiller@cultdeadsheep.org>,Oliver Eikemeier <eikemeier@fillmore-labs.com>
fix a bug in the devel/distcc which becomes obvious when
the port gets built without WITHOUT_GNOME, but when
HAVE_GNOME returns ""
PR: 53705
Submitted by: Frerich Raabe <frerich.raabe@gmx.de> (maintainer)
- A new authentication system for allowing distccd IP-based access
control.
- A new scheduler which can give an up to 10% improvement of
performance of a cluster.
PR: 48665
Submitted by: Frerich Raabe <frerich.raabe@gmx.de>
commands such as .../xgcc -B.../ conftest.C -c -S. Assuming that
xgcc invokes the distcc client, this makes the distcc client write
the assembly output to conftest.o even though it should go into
conftest.s. The cause of this is that distcc currently does not
honour the fact that -S supersedes -c. Interestingly, if you add
"-o conftest.s" to the command line, it's the distcc server that
fails, claiming it couldn't find conftest.C."
o The attached patch fixes this behaviour. Courtesy of [1]
PR: 42019
Submitted by: MAINTAINER,
Alexandre Oliva <aoliva@redhat.com> [1]