On Fri, Oct 10, 2014 at 02:31:55PM +0200, Andrzej Hajda wrote:
Before DPMS off driver disables vblank. It should be balanced by vblank enable after DPMS on. The patch fixes issue with page_flip ioctl not being able to acquire vblank counter introduced by patch: drm: Always reject drm_vblank_get() after drm_vblank_off()
Signed-off-by: Andrzej Hajda a.hajda@samsung.com
Yeah, you should always call vblank_on again when you (re)enable a crtc, whether this is through a set_config call or through dpms. This is Reviewed-by: Daniel Vetter daniel.vetter@ffwll.ch
Sorry that we didn't catch the impact of this additional check on existing drivers, I've thought I've reviewed them and checked that they all call vblank_on. But I didn't take into account that the codepaths might differ for dpms and set_config paths. Otoh most drivers really should implement one in terms of the other.
Cheers, Daniel
drivers/gpu/drm/exynos/exynos_drm_crtc.c | 5 ++++- 1 file changed, 4 insertions(+), 1 deletion(-)
diff --git a/drivers/gpu/drm/exynos/exynos_drm_crtc.c b/drivers/gpu/drm/exynos/exynos_drm_crtc.c index 8e38e9f..45026e6 100644 --- a/drivers/gpu/drm/exynos/exynos_drm_crtc.c +++ b/drivers/gpu/drm/exynos/exynos_drm_crtc.c @@ -71,13 +71,16 @@ static void exynos_drm_crtc_dpms(struct drm_crtc *crtc, int mode) !atomic_read(&exynos_crtc->pending_flip), HZ/20)) atomic_set(&exynos_crtc->pending_flip, 0);
drm_vblank_off(crtc->dev, exynos_crtc->pipe);
drm_crtc_vblank_off(crtc);}
if (manager->ops->dpms) manager->ops->dpms(manager, mode);
exynos_crtc->dpms = mode;
if (mode == DRM_MODE_DPMS_ON)
drm_crtc_vblank_on(crtc);}
static void exynos_drm_crtc_prepare(struct drm_crtc *crtc)
1.9.1
dri-devel mailing list dri-devel@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/dri-devel