[libvirt] [PATCH v3] Avoid a possible NULL pointer dereference in qemuDomainGetTLSObjects

[ashish mittal]

On Wed, Sep 20, 2017 at 6:11 AM, Erik Skultety <eskultet at redhat.com> wrote:

> On Wed, Sep 20, 2017 at 05:32:29AM -0700, Ashish Mittal wrote:
> > Passing a NULL value for the argument secAlias to the function
> > qemuDomainGetTLSObjects would cause a segmentation fault in
> > libvirtd.
> >
> > Changed code to not dereference a NULL secAlias.
> >
> > Signed-off-by: Ashish Mittal <ashmit602 at gmail.com>
> > ---
> >  src/qemu/qemu_hotplug.c | 3 ++-
> >  1 file changed, 2 insertions(+), 1 deletion(-)
> >
> > diff --git a/src/qemu/qemu_hotplug.c b/src/qemu/qemu_hotplug.c
> > index 7dd6e5f..9ecdf0a 100644
> > --- a/src/qemu/qemu_hotplug.c
> > +++ b/src/qemu/qemu_hotplug.c
> > @@ -1643,7 +1643,8 @@ qemuDomainGetTLSObjects(virQEMUCapsPtr qemuCaps,
> >      }
> >
> >      if (qemuBuildTLSx509BackendProps(tlsCertdir, tlsListen, tlsVerify,
> > -                                     *secAlias, qemuCaps, tlsProps) < 0)
> > +                                     secAlias ? *secAlias : NULL,
> qemuCaps,
> > +                                     tlsProps) < 0)
> >          return -1;
> >
> >      if (!(*tlsAlias = qemuAliasTLSObjFromSrcAlias(srcAlias)))
>
> A few lines above this context, we check whether we have a valid reference
> to
> @secinfo and if so, we try to fill the @secAlias. Can we guarantee that
> when
> @secinfo is passed, @secAlias has been set as well?


When secinfo is passed, the line marked below should guarantee that
secAlias is set to not-null.

    if (secinfo) {
        if (qemuBuildSecretInfoProps(secinfo, secProps) < 0)
            return -1;

        if (!(*secAlias = qemuDomainGetSecretAESAlias(srcAlias, false)))
 <=== this should ensure secAlias != NULL or return -1
            return -1;
    }

Please let me know if any changes are required for this patch. Thanks!

Ashish


> I'm asking because I
> haven't touched the TLS code yet and I just ran a few argument combinations
> mentally and this one got me wondering. If the combination I'm describing
> is a
> pure non-sense, the patch can go straight in.
>
> Thanks,
> Erik
>
> > --
> > 2.5.5
> >
> > --
> > libvir-list mailing list
> > libvir-list at redhat.com
> > https://www.redhat.com/mailman/listinfo/libvir-list
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://listman.redhat.com/archives/libvir-list/attachments/20170920/b229ae36/attachment-0001.htm>