Date: Sun, 29 Jul 2012 21:19:58 +0200 From: Luigi Rizzo <rizzo@iet.unipi.it> To: David Chisnall <theraven@FreeBSD.org> Cc: "Bjoern A. Zeeb" <bzeeb-lists@lists.zabbadoz.net>, current@FreeBSD.org Subject: Re: RFC: libkern version of inet_ntoa_r Message-ID: <20120729191958.GB85015@onelab2.iet.unipi.it> In-Reply-To: <CBCC6E24-8D03-422E-B571-1B62FB7667E6@FreeBSD.org> References: <20120725155211.GA33971@onelab2.iet.unipi.it> <alpine.BSF.2.00.1207282213171.4474@ai.fobar.qr> <20120729095833.GB80946@onelab2.iet.unipi.it> <CBCC6E24-8D03-422E-B571-1B62FB7667E6@FreeBSD.org>
next in thread | previous in thread | raw e-mail | index | archive | help
On Sun, Jul 29, 2012 at 05:55:19PM +0100, David Chisnall wrote:
> On 29 Jul 2012, at 10:58, Luigi Rizzo wrote:
>
> > 3. nuke inet_ntoa_r() from libc
>
> inet_ntoa_r is a public symbol and therefore part of our ABI contract with userspace applications. Even if no one that we are aware of is using it, we should officially deprecate it for one major release before removing it. ABI churn for purely aesthetic reasons does not make users happy people.
sure, interpret "nuke" as a long term thing
(starting with deprecation, manpage notes, etc.)
>
> > I need it because i would like to compile parts of the kernel in userspace,
> > and having a kernel function with the same name and different arguments
> > from of a libc function is annoying.
>
>
> Presumably this usage can be trivially fixed with a trivial macro in a prefix header?
Remapping f(a) into f(a, b) requires both a macro
and a wrapping function, something like this
T __f(T1 a, T2 b) { return f(a, b); }
#define f(a) __f(a, b)
Surely can be done (in fact, i have done it already, see
http://info.iet.unipi.it/~luigi/netmap/20120725-ipfw-user.tgz
but i am not so interested in participating to the IOCCC :)
http://www.ioccc.org/
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20120729191958.GB85015>