[libvirt] [PATCH] virprocess: Introduce our own setns() wrapper

[Ian Campbell]

On Mon, 2014-09-15 at 10:21 +0100, Daniel P. Berrange wrote:
> On Sun, Sep 14, 2014 at 06:35:19PM +0100, Ian Campbell wrote:
> > On Wed, 2014-09-10 at 12:20 +0200, Michal Privoznik wrote:
> > >  
> > > +/*
> > > + * Workaround older glibc. While kernel may support the setns
> > > + * syscall, the glibc wrapper might not exist. If that's the
> > > + * case, use our own.
> > > + */
> > > +#ifndef __NR_setns
> > > +# if defined(__x86_64__)
> > > +#  define __NR_setns 308
> > > +# elif defined(__i386__)
> > > +#  define __NR_setns 346
> > > +# else
> > > +#  error "__NR_setns is not defined"
> > > +# endif
> > > +#endif
> > 
> > Xen's automated build tests of libvirt have just failed with this when
> > building for armhf:
> >         util/virprocess.c:75:4: error: #error "__NR_setns is not
> >         defined"
> > 
> > is this another aspect of the "out of date glibc" issue?
> 
> IIUC, the syscalls numbers come from the kernel headers, while the syscall
> API comes from glibc.

Makes sense.

>  Either way, we should not have left this as a fatal
> #error. I'll dig up the required constants for all other architectures we
> care about.

Thanks!

Ian.