Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 2620:137:e000::1:20 as permitted sender) client-ip=2620:137:e000::1:20;
Message-ID: <04cc52e7458cbd22c0ef9d092aabca3ea5403074.camel@kernel.org>
Subject: Re: [PATCH v2] drm: use mgr->dev in drm_dbg_kms in
 drm_dp_add_payload_part2
From:   Jeff Layton <jlayton@kernel.org>
To:     Jani Nikula <jani.nikula@linux.intel.com>,
        David Airlie <airlied@gmail.com>,
        Daniel Vetter <daniel@ffwll.ch>
Cc:     Wayne.Lin@amd.com, lyude@redhat.com, alexdeucher@gmail.com,
        dri-devel@lists.freedesktop.org, linux-kernel@vger.kernel.org
Date:   Sat, 22 Apr 2023 20:34:55 -0400
In-Reply-To: <87o7nkypmo.fsf@intel.com>
References: <20230419112447.18471-1-jlayton@kernel.org>
         <87o7nkypmo.fsf@intel.com>
Content-Type: text/plain; charset="ISO-8859-15"
Content-Transfer-Encoding: quoted-printable
User-Agent: Evolution 3.48.0 (3.48.0-1.fc38) 
MIME-Version: 1.0
Precedence: bulk

On Wed, 2023-04-19 at 16:21 +0300, Jani Nikula wrote:
> On Wed, 19 Apr 2023, Jeff Layton <jlayton@kernel.org> wrote:
> > I've been experiencing some intermittent crashes down in the display
> > driver code. The symptoms are ususally a line like this in dmesg:
> >=20
> >     amdgpu 0000:30:00.0: [drm] Failed to create MST payload for port 00=
0000006d3a3885: -5
> >=20
> > ...followed by an Oops due to a NULL pointer dereference.
> >=20
> > Switch to using mgr->dev instead of state->dev since "state" can be
> > NULL in some cases.
> >=20
> > Link: https://bugzilla.redhat.com/show_bug.cgi?id=3D2184855
> > Suggested-by: Jani Nikula <jani.nikula@linux.intel.com>
> > Signed-off-by: Jeff Layton <jlayton@kernel.org>
>=20
> Thanks,
>=20
> Reviewed-by: Jani Nikula <jani.nikula@intel.com>
>=20
>
> > ---
> >  drivers/gpu/drm/display/drm_dp_mst_topology.c | 2 +-
> >  1 file changed, 1 insertion(+), 1 deletion(-)
> >=20
> > I've been running this patch for a couple of days, but the problem
> > hasn't occurred again as of yet. It seems sane though as long as we can
> > assume that mgr->dev will be valid even when "state" is a NULL pointer.
> >=20
> > diff --git a/drivers/gpu/drm/display/drm_dp_mst_topology.c b/drivers/gp=
u/drm/display/drm_dp_mst_topology.c
> > index 38dab76ae69e..e2e21ce79510 100644
> > --- a/drivers/gpu/drm/display/drm_dp_mst_topology.c
> > +++ b/drivers/gpu/drm/display/drm_dp_mst_topology.c
> > @@ -3404,7 +3404,7 @@ int drm_dp_add_payload_part2(struct drm_dp_mst_to=
pology_mgr *mgr,
> > =20
> >  	/* Skip failed payloads */
> >  	if (payload->vc_start_slot =3D=3D -1) {
> > -		drm_dbg_kms(state->dev, "Part 1 of payload creation for %s failed, s=
kipping part 2\n",
> > +		drm_dbg_kms(mgr->dev, "Part 1 of payload creation for %s failed, ski=
pping part 2\n",
> >  			    payload->port->connector->name);
> >  		return -EIO;
> >  	}
>=20

Thanks for the reviews!

I finally had this happen again today, and I can confirm that this does
prevent the oops. GNOME rearranged my screen layout after the error, but
the box stayed up and running.=20
--=20
Jeff Layton <jlayton@kernel.org>