Re: CVS commit: src/include

[christos%zoulas.com@localhost (Christos Zoulas)]

On Feb 9, 12:15pm, uwe%stderr.spb.ru@localhost (Valery Ushakov) wrote:
-- Subject: Re: CVS commit: src/include

| On Fri, Feb 09, 2018 at 02:44:05 +0000, Christos Zoulas wrote:
| 
| > In article <20180208234417.GA23349%britannica.bec.de@localhost>,
| > Joerg Sonnenberger  <joerg%bec.de@localhost> wrote:
| > >On Thu, Feb 08, 2018 at 10:56:22PM +0000, Christos Zoulas wrote:
| > >> In article <20180207130259.GA21472%britannica.bec.de@localhost>,
| > >> Joerg Sonnenberger  <joerg%bec.de@localhost> wrote:
| > >> >On Tue, Feb 06, 2018 at 03:21:21PM -0500, Christos Zoulas wrote:
| > >> >> Module Name:	src
| > >> >> Committed By:	christos
| > >> >> Date:		Tue Feb  6 20:21:21 UTC 2018
| > >> >> 
| > >> >> Modified Files:
| > >> >> 	src/include: unistd.h
| > >> >> 
| > >> >> Log Message:
| > >> >> detect duplicate declaration of pthread_atfork() in pthread.h
| > >> >
| > >> >Is this for some new broken GCC warning?
| > >> >
| > >> 
| > >> This is what the compiler said, and we have prior art for this (macro
| > >> protect to avoid dup declarations -- grep for _DECLARED).
| > >
| > >Normally only for typedefs, since pre-C11 (?) it was invalid to typedef
| > >the same thing twice. That's not true for prototypes though.
| > 
| > We have been doing this for functions for a while; check unistd.h
| 
| But *why* have we been doing that?

I added the redundand decls warning in bsd.sys.mk in 2001,
but it is still commented out. Is that now part of -Wall?

christos