On Mon, May 18, 2020 at 2:33 PM Stephen Smalley
<stephen.smalley.work@xxxxxxxxx> wrote:
>
> On Mon, May 18, 2020 at 2:14 PM James Carter <jwcart2@xxxxxxxxx> wrote:
> >
> > Initialize the multiple_decls field when intializing the structure
> > cil_db.
> >
> > Issue reported by: Topi Miettinen <toiwoton@xxxxxxxxx>
> >
> > Fixes: fafe4c212bf6c32c ("libsepol: cil: Add ability to redeclare
> > types[attributes]")
>
> I guess this explains the non-deterministic behavior and valgrind
> warning about uninitialized data, but what about the segfault?
> Don't we still need to fix the case when multiple_decls is set to TRUE?
Yes. I am working on that. I believe the segfault occurs when you
declare something both a type and a typeattribute.
Jim
>
> >
> > Signed-off-by: James Carter <jwcart2@xxxxxxxxx>
> > ---
> > libsepol/cil/src/cil.c | 1 +
> > 1 file changed, 1 insertion(+)
> >
> > diff --git a/libsepol/cil/src/cil.c b/libsepol/cil/src/cil.c
> > index 4a77aa9c..a3c6a293 100644
> > --- a/libsepol/cil/src/cil.c
> > +++ b/libsepol/cil/src/cil.c
> > @@ -459,6 +459,7 @@ void cil_db_init(struct cil_db **db)
> > (*db)->preserve_tunables = CIL_FALSE;
> > (*db)->handle_unknown = -1;
> > (*db)->mls = -1;
> > + (*db)->multiple_decls = CIL_FALSE;
> > (*db)->target_platform = SEPOL_TARGET_SELINUX;
> > (*db)->policy_version = POLICYDB_VERSION_MAX;
> > }
> > --
> > 2.25.4
