On Mon, May 6, 2019 at 8:49 AM <dufa...@hda.com> wrote: > > > On May 6, 2019, at 01:02 , Sebastian Huber < > sebastian.hu...@embedded-brains.de> wrote: > > On 05/05/2019 03:00, Chris Johns wrote: > > On 5/5/19 12:21 am, Joel Sherrill wrote: > > Sorry for missing that it actually was referenced > > Rcmd can and should work. Rcmdsh will not ever and should be removed from > libbsd. > > Does rcmd need testing? It clearly had no test since it wouldn't have > linked. > > > Yes that would be great. Testing libbsd is a complicated. > > > If it is used by applications it should be tested, otherwise we should > remove it from the build to make it a little faster and reduce the > maintenance costs of libbsd. > > > It is used. The use may be removed (this is a port from vxWorks) but as > part of a baseline it is used. >
Just to be specific again. rcmd() can be made to work. rcmdsh() uses fork() and exec() so will never work. Lou's patch added conditionals to rcmd() to get around missing netgroup methods and avoid invoking rcmdsh() on one path. An update to this patch (or a follow up) needs to remove rcmdsh* from the rtems-libbsd source I suppose Lou also needs to add a test that verifies rcmd() works. --joel > > > Peter > ----------------- > Peter Dufault > HD Associates, Inc. Software and System Engineering > > This email is delivered through the public internet using protocols > subject to interception and tampering. > >
_______________________________________________ devel mailing list devel@rtems.org http://lists.rtems.org/mailman/listinfo/devel
