On Wed, Nov 26, 2014 at 06:58:04PM -0500, Rob Clark wrote:
> Otherwise we'd still end up w/ the plane attached to the CRTC, and
> seemingly active, but without an FB. Which ends up going *boom*
> in the drivers.
>
> Slightly modified version of Daniel's irc suggestion.
>
> CC: Daniel Vetter <daniel at ffwll.ch>
> Signed-off-by: Rob Clark <robdclark at gmail.com>
I think the problem here really is just that legacy state (plane->fb/crtc)
and atomic state get out of sync. Leaving planes around while the crtc
goes down is already possible with ->disable_plane and overlays, drivers
should be able to cope with that.
Merged into my atomic fixes pile, thanks.
-Daniel
> ---
> drivers/gpu/drm/drm_atomic_helper.c | 19 +++++++++++++------
> 1 file changed, 13 insertions(+), 6 deletions(-)
>
> diff --git a/drivers/gpu/drm/drm_atomic_helper.c
> b/drivers/gpu/drm/drm_atomic_helper.c
> index a348688..7d5e6c1 100644
> --- a/drivers/gpu/drm/drm_atomic_helper.c
> +++ b/drivers/gpu/drm/drm_atomic_helper.c
> @@ -1409,11 +1409,24 @@ retry:
> goto fail;
> }
>
> + primary_state = drm_atomic_get_plane_state(state, crtc->primary);
> + if (IS_ERR(primary_state)) {
> + ret = PTR_ERR(primary_state);
> + goto fail;
> + }
> +
> if (!set->mode) {
> WARN_ON(set->fb);
> WARN_ON(set->num_connectors);
>
> crtc_state->enable = false;
> +
> + ret = drm_atomic_set_crtc_for_plane(state, crtc->primary, NULL);
> + if (ret != 0)
> + goto fail;
> +
> + drm_atomic_set_fb_for_plane(primary_state, NULL);
> +
> goto commit;
> }
>
> @@ -1423,12 +1436,6 @@ retry:
> crtc_state->enable = true;
> drm_mode_copy(&crtc_state->mode, set->mode);
>
> - primary_state = drm_atomic_get_plane_state(state, crtc->primary);
> - if (IS_ERR(primary_state)) {
> - ret = PTR_ERR(primary_state);
> - goto fail;
> - }
> -
> ret = drm_atomic_set_crtc_for_plane(state, crtc->primary, crtc);
> if (ret != 0)
> goto fail;
> --
> 1.9.3
>
--
Daniel Vetter
Software Engineer, Intel Corporation
+41 (0) 79 365 57 48 - http://blog.ffwll.ch
