Inetd.c sigpause() --> __xpg_sigpause() not in uClibc

John Z. Bohach jzb at aexorsyst.com
Wed Nov 9 01:57:43 UTC 2005


On Tuesday 08 November 2005 17:20, Rob Landley wrote:
> On Tuesday 08 November 2005 17:40, John Z. Bohach wrote:
> > The recent synching of inetd.c with the Slackware version has the
> > undesirable sideffect of using sigpause() (network/inetd.c, line 1386).
> >
> > According to the sigpause() manpage, sigpause is deprecated and
> > sigsuspend() should be used instead.
>
> A patch to add usage of a deprecated ioctl is broken, yes.  Care to send a
> tested patch?

Its not an ioctl().  Its a standard C library routine call.  The interesting 
thing is that this WAS fine:  i.e., there was no 'sigpause()' in the rev. 
11010 of network/inetd.c, but the glorious slackware synch'up replaced known 
good code with slackware code which in this case breaks uClibc compatibility.  
More interesting to know is am I the first one to notice this, or has noone 
tried a buildroot with today's snapshot since oct. 3, 2005, which is when 
inetd.c was changed to slackware's inetd.c (that's when sigpause() was 
added).

As for a patch, I can revert those few lines to the ones in rev. 11010, as I 
don't think those were the ones that caused the slackware synch'up to be 
necessary, but it would be more appropriate if the person(s) who modified it 
and commented "sync inetd with slackware-10.2 inetd (OpenBSD-1.79), have 
destroy bug 455" as well as the person(s) who patched it later in following 
revs. think that's okay for a long-term solution to put that part of the file 
back the way it was.

--john

-- 
     ###  Any similarity between my views and the truth is completely ###
     ###  coincidental, except that they are endorsed by NO ONE       ###




More information about the busybox mailing list