On Tue, Jan 07, 2025 at 03:36:20PM +0100, Maxime Ripard wrote:
> On Wed, Dec 25, 2024 at 01:10:13AM +0200, Dmitry Baryshkov wrote:
> > CEC adapter functions are protected by drm_dev_enter() / _exit(), so it
> > is safe to keep CEC registered until DRM device is being torn down. Drop
> > custom CEC unsregistration code and rely on the DRM core to unregister
> > CEC adapter.
> >
> > Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@xxxxxxxxxx>
> > ---
> > drivers/gpu/drm/vc4/vc4_hdmi.c | 53 ++++++------------------------------------
> > 1 file changed, 7 insertions(+), 46 deletions(-)
> >
> > diff --git a/drivers/gpu/drm/vc4/vc4_hdmi.c b/drivers/gpu/drm/vc4/vc4_hdmi.c
> > index 3086c2ad3bb2e8fafdc1f37ba985aa5785d49f9a..7b39000369e50f5368384575aa81919df5a7133c 100644
> > --- a/drivers/gpu/drm/vc4/vc4_hdmi.c
> > +++ b/drivers/gpu/drm/vc4/vc4_hdmi.c
> > @@ -2752,13 +2752,6 @@ static const struct cec_adap_ops vc4_hdmi_cec_adap_ops = {
> > .adap_transmit = vc4_hdmi_cec_adap_transmit,
> > };
> >
> > -static void vc4_hdmi_cec_release(void *ptr)
> > -{
> > - struct drm_connector *connector = ptr;
> > -
> > - drm_connector_cec_unregister(connector);
> > -}
> > -
> > static int vc4_hdmi_cec_init(struct drm_connector *connector)
> > {
> > struct vc4_hdmi *vc4_hdmi = connector_to_vc4_hdmi(connector);
> > @@ -2796,51 +2789,19 @@ static int vc4_hdmi_cec_register(struct vc4_hdmi *vc4_hdmi)
> > {
> > struct platform_device *pdev = vc4_hdmi->pdev;
> > struct device *dev = &pdev->dev;
> > - int ret;
> >
> > if (!of_property_present(dev->of_node, "interrupts")) {
> > dev_warn(dev, "'interrupts' DT property is missing, no CEC\n");
> > return 0;
> > }
> >
> > - ret = drm_connector_hdmi_cec_adapter_register(&vc4_hdmi->connector,
> > - &vc4_hdmi_cec_adap_ops,
> > - vc4_hdmi->variant->card_name,
> > - 1,
> > - vc4_hdmi_cec_init,
> > - NULL,
> > - &pdev->dev);
> > - if (ret < 0)
> > - return ret;
> > -
> > - /*
> > - * NOTE: Strictly speaking, we should probably use a DRM-managed
> > - * registration there to avoid removing the CEC adapter by the
> > - * time the DRM driver doesn't have any user anymore.
> > - *
> > - * However, the CEC framework already cleans up the CEC adapter
> > - * only when the last user has closed its file descriptor, so we
> > - * don't need to handle it in DRM.
> > - *
> > - * By the time the device-managed hook is executed, we will give
> > - * up our reference to the CEC adapter and therefore don't
> > - * really care when it's actually freed.
> > - *
> > - * There's still a problematic sequence: if we unregister our
> > - * CEC adapter, but the userspace keeps a handle on the CEC
> > - * adapter but not the DRM device for some reason. In such a
> > - * case, our vc4_hdmi structure will be freed, but the
> > - * cec_adapter structure will have a dangling pointer to what
> > - * used to be our HDMI controller. If we get a CEC call at that
> > - * moment, we could end up with a use-after-free. Fortunately,
> > - * the CEC framework already handles this too, by calling
> > - * cec_is_registered() in cec_ioctl() and cec_poll().
> > - */
>
> I'd really like to keep that comment, somewhere.
Ack, I will try to rephrase it so that it still fits the changed source.
--
With best wishes
Dmitry
