On Mon, Jul 10, 2017 at 16:07:08 +0100, Daniel Berrange wrote:
On Mon, Jul 10, 2017 at 01:01:26PM +0200, Peter Krempa wrote:
> On Wed, Jul 05, 2017 at 12:58:49 +0100, Daniel Berrange wrote:
> > The HOST_NAME_MAX, INET_ADDRSTRLEN and VIR_LOOPBACK_IPV4_ADDR
> > constants are only used by a handful of files, so are better
> > kept in virsocketaddr.h
> >
> > Signed-off-by: Daniel P. Berrange <berrange(a)redhat.com>
> > ---
> > src/internal.h | 16 ----------------
> > src/libxl/libxl_conf.c | 1 +
> > src/nwfilter/nwfilter_dhcpsnoop.c | 1 +
> > src/nwfilter/nwfilter_gentech_driver.c | 1 +
> > src/qemu/qemu_conf.c | 1 +
> > src/util/virsocketaddr.h | 16 ++++++++++++++++
> > src/util/virutil.c | 1 +
> > src/vz/vz_sdk.c | 1 +
> > 8 files changed, 22 insertions(+), 16 deletions(-)
> >
>
> [...]
>
> > diff --git a/src/util/virsocketaddr.h b/src/util/virsocketaddr.h
> > index 43a3706..ae76166 100644
> > --- a/src/util/virsocketaddr.h
> > +++ b/src/util/virsocketaddr.h
> > @@ -32,6 +32,22 @@
> > # include <sys/un.h>
> > # endif
> >
> > +/* On architectures which lack these limits, define them (ie. Cygwin).
> > + * Note that the libvirt code should be robust enough to handle the
> > + * case where actual value is longer than these limits (eg. by setting
> > + * length correctly in second argument to gethostname and by always
> > + * using strncpy instead of strcpy).
> > + */
> > +# ifndef HOST_NAME_MAX
> > +# define HOST_NAME_MAX 256
>
> I'm not entirely convinced that this semantically belongs to
> virsocketaddr.h
>
> At least in the virlog.c usage case it is not used in any way in regards
> to network communication.
The constant is used to define a hostname string buffer, which is then
used with the gethostname() system call. That's network related IMHO.
I find it weird, that it gets defined in virsocketaddr.h, but the only
usage is in src/util/virutil.c thus it needs cross-inclusions which were
not in place prior to the move.
Should perhaps virGetHostnameImpl and friends be moved too in that case?