drm/i915: Do a better job at disabling primary plane in the noatomic case.
authorMaarten Lankhorst <maarten.lankhorst@linux.intel.com>
Mon, 23 Nov 2015 09:25:28 +0000 (10:25 +0100)
committerJani Nikula <jani.nikula@intel.com>
Thu, 10 Dec 2015 11:33:42 +0000 (13:33 +0200)
commit634b3a4a476e96816d5d6cd5bb9f8900a53f56ba
tree8afc5f90bd948c6ca3c36c868ab0f2b63eb4a3e4
parent6704d45528537ea6088aeea0667d87b605b82d51
drm/i915: Do a better job at disabling primary plane in the noatomic case.

When disable_noatomic is called plane_mask is not correct yet, and
plane_state->visible = true is left as true after disabling the primary
plane.

Other planes are already disabled as part of crtc sanitization, only the
primary is left active. But the plane_mask is not updated here. It gets
updated during fb takeover in modeset_gem_init, or set to the new value
on resume.

This means that to disable the primary plane 1 << drm_plane_index(primary)
needs to be used.

Afterwards because the crtc is no longer active it's forbidden to keep
plane_state->visible set, or a WARN_ON in
intel_plane_atomic_calc_changes triggers. There are other code points
that rely on accurate plane_state->visible too, so make sure the bool is
cleared.

The other planes are already disabled in intel_sanitize_crtc, so they
don't have to be handled here.

Cc: stable@vger.kernel.org #v4.3, v4.2?
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=92655
Tested-by: Tomas Mezzadra <tmezzadra@gmail.com>
Signed-off-by: Maarten Lankhorst <maarten.lankhorst@linux.intel.com>
Reviewed-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Link: http://patchwork.freedesktop.org/patch/msgid/5652DB88.9070208@linux.intel.com
(cherry picked from commit 54a4196188eab82e6f0a5f05716626e9f18b8fb6)
Signed-off-by: Jani Nikula <jani.nikula@intel.com>
drivers/gpu/drm/i915/intel_display.c