Am Mittwoch, den 19.10.2016, 15:12 +0800 schrieb Ying Liu: > On Wed, Oct 19, 2016 at 12:07 AM, Philipp Zabel <p.zabel@xxxxxxxxxxxxxx> wrote: > > Odd x/y offsets are not allowed for chroma subsampled planar YUV > > formats. > > > > Signed-off-by: Philipp Zabel <p.zabel@xxxxxxxxxxxxxx> > > --- > > drivers/gpu/drm/imx/ipuv3-plane.c | 8 ++++++++ > > 1 file changed, 8 insertions(+) > > > > diff --git a/drivers/gpu/drm/imx/ipuv3-plane.c b/drivers/gpu/drm/imx/ipuv3-plane.c > > index 5c34299..3a03fd8 100644 > > --- a/drivers/gpu/drm/imx/ipuv3-plane.c > > +++ b/drivers/gpu/drm/imx/ipuv3-plane.c > > @@ -259,6 +259,7 @@ static int ipu_plane_atomic_check(struct drm_plane *plane, > > struct drm_framebuffer *fb = state->fb; > > struct drm_framebuffer *old_fb = old_state->fb; > > unsigned long eba, ubo, vbo, old_ubo, old_vbo; > > + int hsub, vsub; > > > > /* Ok to disable */ > > if (!fb) > > @@ -372,6 +373,13 @@ static int ipu_plane_atomic_check(struct drm_plane *plane, > > > > if (old_fb && old_fb->pitches[1] != fb->pitches[1]) > > crtc_state->mode_changed = true; > > + > > + /* x and y offsets must be even in case of chroma subsampling */ > > True for x offsets, but not for y offsets, e.g., NV16. That should be ok, for NV16 vsub==1, so (y&(vsub-1)) will always be false. Only for 4:2:0 subsampled formats is vsub==2, and then (y&(vsub-1)) will only be true if y is odd. I'll change the comment to: /* * The x/y offsets must be even in case of horizontal/vertical * chroma subsampling. */ regards Philipp _______________________________________________ dri-devel mailing list dri-devel@xxxxxxxxxxxxxxxxxxxxx https://lists.freedesktop.org/mailman/listinfo/dri-devel
