[Date Prev][Date Next]   [Thread Prev][Thread Next]   [Thread Index] [Date Index] [Author Index]

Re: rpmlint message '%ifarch-applied-patch'



On Wed, Sep 19, 2007 at 01:04:18PM +0100, Richard W.M. Jones wrote:
> What does this rpmlint message mean?  I don't understand why it's 
> harmful to have patches only applying to particular architectures.

It is harmful, because the patch isn't tested on all the other arches.
Also eventually if further patches are applied which touch the same
locations you end up with a twisty unmaintanable maze of per-arch
patches.

> The patch in question is:

That is very bogus patch.
1) gethostbyname has the second argument size_t, so if anything,
   you should patch it everywhere for the right type
2) you really should find out why configure doesn't find
   gethostname prototype on Linux, it is present in unistd.h
   in glibc on all arches, unless you request a namespace which
   forbids it in unistd.h.

> --- ical-2.2/calendar/uid.C.ia64	Wed May 29 22:51:03 1996
> +++ ical-2.2/calendar/uid.C	Mon Apr 30 00:13:58 2001
> @@ -13,7 +13,7 @@
>  #ifndef HAVE_GETHOSTNAME_PROTO
> 
>  #ifdef linux
> -extern "C" int gethostname(char*, unsigned int);
> +extern "C" int gethostname(char*, unsigned long);
>  #else /* !linux */
>  extern "C" int gethostname(char*, int);
>  #endif /* linux */
> 

	Jakub


[Date Prev][Date Next]   [Thread Prev][Thread Next]   [Thread Index] [Date Index] [Author Index]