Date: Wed, 13 Apr 2005 10:45:04 -0600 From: Scott Long <scottl@samsco.org> To: "M. Warner Losh" <imp@bsdimp.com> Cc: davidxu@freebsd.org Subject: Re: HEADS-UP: Planning on deprecating libc_r for 6.0 Message-ID: <425D4C90.9090702@samsco.org> In-Reply-To: <20050413.085652.56562781.imp@bsdimp.com> References: <425D2041.7020501@samsco.org> <20050413.085652.56562781.imp@bsdimp.com>
next in thread | previous in thread | raw e-mail | index | archive | help
M. Warner Losh wrote: > In message: <425D2041.7020501@samsco.org> > Scott Long <scottl@samsco.org> writes: > : One question that has come up is how to warn the user at runtime about > : this deprecation. Should the dynamic linker print a message to stderr > : when it gets a request to load libc_r? Should it go to the console > : and/or syslog instead? Should there be a way to disable these messages > : so as not to break wrapper programs that might be confused by the > : output? Should we even bother at all with runtime warnings? > > I'd do a build-time warning only. > > This would allow poeple to deploy systems with libc_r and not have > annoying warnings at a customer's site. They will know that things > are deprecated. Also, compat librarires will be around for a while, > so doing things at runtime seems annoying... > > As far as my company's use of libc_r, we don't yet know how well > non-libc_r works yet since we've not done any large threaded FreeBSD 5 > applications yet. So I can't comment on how wise/foolish this option > is with any direct experience. > > Warner Yes, outfits like yours and Julian's are exactly what I had in mind when I said that it would be several years before it goes away, and even then it would live on in /usr/lib/compat. Scott
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?425D4C90.9090702>