On Wed, Feb 03, 2016 at 10:34:06AM +0000, Richard W.M. Jones wrote:
On Wed, Feb 03, 2016 at 10:35:07AM +0100, Pino Toscano wrote:
> On Tuesday 02 February 2016 19:47:12 Richard W.M. Jones wrote:
> > On Tue, Feb 02, 2016 at 03:27:39PM +0100, Pino Toscano wrote:
> > > diff --git a/src/launch.c b/src/launch.c
> > > index f59818f..ec061e3 100644
> > > --- a/src/launch.c
> > > +++ b/src/launch.c
> > > @@ -418,6 +418,21 @@ guestfs_int_get_cpu_model (int kvm)
> > > #endif
> > > }
> > >
> > > +/* Create the path for a socket with the selected filename in the
> > > + * tmpdir.
> > > + */
> > > +int
> > > +guestfs_int_create_socketname (guestfs_h *g, const char *filename,
> > > + char (*sockpath)[UNIX_PATH_MAX])
> > > +{
> > > + char *path = g->tmpdir;
> > > +
> > > + snprintf (*sockpath, UNIX_PATH_MAX-1, "%s/%s", path,
filename);
> > > + (*sockpath)[UNIX_PATH_MAX-1] = '\0';
> >
> > What's wrong with:
> >
> > snprintf (*sockpath, UNIX_PATH_MAX, "%s/%s", path, filename);
>
> If the "$path/$filename" string is longer than UNIX_PATH_MAX, then
> *sockpath won't be 0-terminated.
That's not true though:
Having said that, what this function _could_ usefully do is to give a
fatal error if the final path is longer than UNIX_PATH_MAX-1, since
truncating the path is unlikely to produce correct behaviour.
Rich.
--
Richard Jones, Virtualization Group, Red Hat
http://people.redhat.com/~rjones
Read my programming and virtualization blog:
http://rwmj.wordpress.com
virt-builder quickly builds VMs from scratch
http://libguestfs.org/virt-builder.1.html