This reverts commit 13803132818cf8084d169617be060fd8e3411a98.
This patch need to be revert because it breaks middlewares way of working. As example, modetest and weston, only relies on drmModeRmFB to close CRTC or planes. Keeping this patch will induce weird behavior like a plane displayed with modetest will be visible on weston and vice-versa.
We can overcome this by updating the middleware (successfully tested with modetest) to force them to disable CRTC and plane using drmModeSetCrtc or drmModeSetPlane. But it is a long to way and for short term, it is not acceptable to break ABI.
Cc: Maarten Lankhorst maarten.lankhorst@linux.intel.com Cc: David Herrmann dh.herrmann@gmail.com Cc: Daniel Vetter daniel.vetter@ffwll.ch Signed-off-by: Vincent Abriou vincent.abriou@st.com --- drivers/gpu/drm/drm_crtc.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-)
diff --git a/drivers/gpu/drm/drm_crtc.c b/drivers/gpu/drm/drm_crtc.c index 626b0a5..9b9c4b4 100644 --- a/drivers/gpu/drm/drm_crtc.c +++ b/drivers/gpu/drm/drm_crtc.c @@ -3327,7 +3327,7 @@ int drm_mode_rmfb(struct drm_device *dev, mutex_unlock(&dev->mode_config.fb_lock); mutex_unlock(&file_priv->fbs_lock);
- drm_framebuffer_unreference(fb); + drm_framebuffer_remove(fb);
return 0;
@@ -3517,7 +3517,7 @@ void drm_fb_release(struct drm_file *priv) list_del_init(&fb->filp_head);
/* This will also drop the fpriv->fbs reference. */ - drm_framebuffer_unreference(fb); + drm_framebuffer_remove(fb); } }