Date: Mon, 27 Oct 2008 16:34:02 -0400 From: John Baldwin <jhb@freebsd.org> To: Ed Schouten <ed@80386.nl> Cc: svn-src-stable@freebsd.org, svn-src-all@freebsd.org, src-committers@freebsd.org, svn-src-stable-7@freebsd.org Subject: Re: svn commit: r184300 - in stable/7/lib: libc/stdlib libutil Message-ID: <200810271634.03328.jhb@freebsd.org> In-Reply-To: <20081027194008.GO6808@hoeg.nl> References: <200810262155.m9QLtJG5096815@svn.freebsd.org> <200810271422.06751.jhb@freebsd.org> <20081027194008.GO6808@hoeg.nl>
next in thread | previous in thread | raw e-mail | index | archive | help
On Monday 27 October 2008 03:40:08 pm Ed Schouten wrote: > * John Baldwin <jhb@freebsd.org> wrote: > > I would perhaps add a note that the duplicate revoke() in openpty() is only to > > support legacy libc's with broken unlockpt() routines. We could maybe remove > > the revoke()/ptsname() from openpty() on 8.x though as all 8.x machines > > should have a working unlockpt(). > > Good point, but I'd rather leave revoke() there for at least a couple of > months. If people just download the openpty() source from -CURRENT > through cvsweb and use it as an example for their own application, they > could create a potential security issue when they run the application on > RELENG_*. > > Shall we leave the revoke() call there for now, but remove it before we > ship 8.0-RELEASE? I would go ahead and axe it from 8 now since the safety net bits are in 6.x and 7.x already. I honestly wouldn't expect people to use openpty()'s implementation as the reference way to use posix_openpt() and friends. Rather, I imagine they would derive that from manpages online or other sources. -- John Baldwin
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?200810271634.03328.jhb>