| Message ID | 20170927193209.11870-1-eric@anholt.net (mailing list archive) |
|---|---|
| State | New, archived |
| Headers | show |
On Wed, 27 Sep 2017 12:32:09 -0700 Eric Anholt <eric@anholt.net> wrote: > The documentation said to use src_w here, and I didn't consider that > we actually needed to be using pitch somewhere in our setup. Fixes > scanout on my DSI panel when X11 does initial setup with 1920x1080 > HDMI and 800x480 DSI both at 0,0 of the same framebuffer. > > Signed-off-by: Eric Anholt <eric@anholt.net> > Fixes: 98830d91da08 ("drm/vc4: Add T-format scanout support.") > --- > drivers/gpu/drm/vc4/vc4_plane.c | 20 +++++++++++++++----- > 1 file changed, 15 insertions(+), 5 deletions(-) > > diff --git a/drivers/gpu/drm/vc4/vc4_plane.c b/drivers/gpu/drm/vc4/vc4_plane.c > index 2968b3ebb895..4ad0b9fcae99 100644 > --- a/drivers/gpu/drm/vc4/vc4_plane.c > +++ b/drivers/gpu/drm/vc4/vc4_plane.c > @@ -547,14 +547,24 @@ static int vc4_plane_mode_set(struct drm_plane *plane, > tiling = SCALER_CTL0_TILING_LINEAR; > pitch0 = VC4_SET_FIELD(fb->pitches[0], SCALER_SRC_PITCH); > break; > - case DRM_FORMAT_MOD_BROADCOM_VC4_T_TILED: > + > + case DRM_FORMAT_MOD_BROADCOM_VC4_T_TILED: { > + /* For T-tiled, the FB pitch is "how many bytes from > + * one row to the next, such that pitch * tile_h == > + * tile_size * tiles_per_row." > + */ > + u32 tile_size_shift = 12; > + u32 tile_h_shift = 5; Maybe you should explain where those _shift values come from in the above comment (tile size is 4K hence tile_size_shift = 12 and a tile is a 32x32 pixels element, hence tile_h_shift = 5). Other than that, Reviewed-by: Boris Brezillon <boris.brezillon@free-electrons.com> > + u32 tiles_w = fb->pitches[0] >> (tile_size_shift - tile_h_shift); > + > tiling = SCALER_CTL0_TILING_256B_OR_T; > > - pitch0 = (VC4_SET_FIELD(0, SCALER_PITCH0_TILE_Y_OFFSET), > - VC4_SET_FIELD(0, SCALER_PITCH0_TILE_WIDTH_L), > - VC4_SET_FIELD((vc4_state->src_w[0] + 31) >> 5, > - SCALER_PITCH0_TILE_WIDTH_R)); > + pitch0 = (VC4_SET_FIELD(0, SCALER_PITCH0_TILE_Y_OFFSET) | > + VC4_SET_FIELD(0, SCALER_PITCH0_TILE_WIDTH_L) | > + VC4_SET_FIELD(tiles_w, SCALER_PITCH0_TILE_WIDTH_R)); > break; > + } > + > default: > DRM_DEBUG_KMS("Unsupported FB tiling flag 0x%16llx", > (long long)fb->modifier);
diff --git a/drivers/gpu/drm/vc4/vc4_plane.c b/drivers/gpu/drm/vc4/vc4_plane.c index 2968b3ebb895..4ad0b9fcae99 100644 --- a/drivers/gpu/drm/vc4/vc4_plane.c +++ b/drivers/gpu/drm/vc4/vc4_plane.c @@ -547,14 +547,24 @@ static int vc4_plane_mode_set(struct drm_plane *plane, tiling = SCALER_CTL0_TILING_LINEAR; pitch0 = VC4_SET_FIELD(fb->pitches[0], SCALER_SRC_PITCH); break; - case DRM_FORMAT_MOD_BROADCOM_VC4_T_TILED: + + case DRM_FORMAT_MOD_BROADCOM_VC4_T_TILED: { + /* For T-tiled, the FB pitch is "how many bytes from + * one row to the next, such that pitch * tile_h == + * tile_size * tiles_per_row." + */ + u32 tile_size_shift = 12; + u32 tile_h_shift = 5; + u32 tiles_w = fb->pitches[0] >> (tile_size_shift - tile_h_shift); + tiling = SCALER_CTL0_TILING_256B_OR_T; - pitch0 = (VC4_SET_FIELD(0, SCALER_PITCH0_TILE_Y_OFFSET), - VC4_SET_FIELD(0, SCALER_PITCH0_TILE_WIDTH_L), - VC4_SET_FIELD((vc4_state->src_w[0] + 31) >> 5, - SCALER_PITCH0_TILE_WIDTH_R)); + pitch0 = (VC4_SET_FIELD(0, SCALER_PITCH0_TILE_Y_OFFSET) | + VC4_SET_FIELD(0, SCALER_PITCH0_TILE_WIDTH_L) | + VC4_SET_FIELD(tiles_w, SCALER_PITCH0_TILE_WIDTH_R)); break; + } + default: DRM_DEBUG_KMS("Unsupported FB tiling flag 0x%16llx", (long long)fb->modifier);
The documentation said to use src_w here, and I didn't consider that we actually needed to be using pitch somewhere in our setup. Fixes scanout on my DSI panel when X11 does initial setup with 1920x1080 HDMI and 800x480 DSI both at 0,0 of the same framebuffer. Signed-off-by: Eric Anholt <eric@anholt.net> Fixes: 98830d91da08 ("drm/vc4: Add T-format scanout support.") --- drivers/gpu/drm/vc4/vc4_plane.c | 20 +++++++++++++++----- 1 file changed, 15 insertions(+), 5 deletions(-)