On Tue, Apr 21 2020 at 2:32P -0400,
Dmitry Baryshkov <dbaryshkov@xxxxxxxxx> wrote:
> вт, 21 апр. 2020 г. в 21:27, Mike Snitzer <snitzer@xxxxxxxxxx>:
> >
> > On Mon, Apr 20 2020 at 9:46P -0400,
> > Dmitry Baryshkov <dbaryshkov@xxxxxxxxx> wrote:
> >
> > > From: Dmitry Baryshkov <dmitry_baryshkov@xxxxxxxxxx>
> > >
> > > Allow one to use encrypted in addition to user and login key types for
> > > device encryption.
> > >
> > > Signed-off-by: Dmitry Baryshkov <dmitry_baryshkov@xxxxxxxxxx>
> >
> > I fixed up some issues, please see the following incremental patch,
> > I'll get this folded in and staged for 5.8.
>
> Thank you!
Actually, I think this is needed too -- but please correct me if I'm wrong:
diff --git a/drivers/md/dm-crypt.c b/drivers/md/dm-crypt.c
index a4c4afc67a3d..ba4d15476862 100644
--- a/drivers/md/dm-crypt.c
+++ b/drivers/md/dm-crypt.c
@@ -2235,8 +2235,9 @@ static int set_key_user(struct crypt_config *cc, struct key *key)
static int set_key_encrypted(struct crypt_config *cc, struct key *key)
{
- struct encrypted_key_payload *ekp = key->payload.data[0];
+ const struct encrypted_key_payload *ekp;
+ ekp = dereference_key_locked(key);
if (!ekp)
return -EKEYREVOKED;
--
dm-devel mailing list
dm-devel@xxxxxxxxxx
https://www.redhat.com/mailman/listinfo/dm-devel
