Re: [PATCH v2 3/9] drm/bridge: ite-it6263: handle unsupported InfoFrames
From: Liu Ying <victor.liu@nxp.com>
Date: 2025-09-29 07:55:06
Also in:
dri-devel, linux-arm-msm, linux-rockchip, linux-sunxi, lkml
On 09/28/2025, Dmitry Baryshkov wrote:
quoted hunk ↗ jump to hunk
Make hdmi_write_hdmi_infoframe() and hdmi_clear_infoframe() callbacks return -EOPNOTSUPP for unsupported InfoFrames and make sure that atomic_check() callback doesn't allow unsupported InfoFrames to be enabled. Signed-off-by: Dmitry Baryshkov <redacted> --- drivers/gpu/drm/bridge/ite-it6263.c | 27 +++++++++++++++++++++++++-- 1 file changed, 25 insertions(+), 2 deletions(-)diff --git a/drivers/gpu/drm/bridge/ite-it6263.c b/drivers/gpu/drm/bridge/ite-it6263.c index 2eb8fba7016cbf0dcb19aec4ca8849f1fffaa64c..cf3d76d748dde51e93b2b19cc2cbe023ca2629b8 100644 --- a/drivers/gpu/drm/bridge/ite-it6263.c +++ b/drivers/gpu/drm/bridge/ite-it6263.c@@ -26,6 +26,7 @@ #include <drm/drm_crtc.h> #include <drm/drm_edid.h> #include <drm/drm_of.h> +#include <drm/drm_print.h> #include <drm/drm_probe_helper.h> /* -----------------------------------------------------------------------------@@ -772,7 +773,7 @@ static int it6263_hdmi_clear_infoframe(struct drm_bridge *bridge, regmap_write(it->hdmi_regmap, HDMI_REG_PKT_NULL_CTRL, 0); break; default: - dev_dbg(it->dev, "unsupported HDMI infoframe 0x%x\n", type); + return -EOPNOTSUPP; } return 0;@@ -812,13 +813,35 @@ static int it6263_hdmi_write_infoframe(struct drm_bridge *bridge, ENABLE_PKT | REPEAT_PKT); break; default: - dev_dbg(it->dev, "unsupported HDMI infoframe 0x%x\n", type); + return -EOPNOTSUPP; } return 0; } +static int it6263_bridge_atomic_check(struct drm_bridge *bridge, + struct drm_bridge_state *bridge_state, + struct drm_crtc_state *crtc_state, + struct drm_connector_state *conn_state) +{ + /* not supported by the driver */ + conn_state->hdmi.infoframes.spd.set = false; + + /* should not happen, audio support not enabled */ + if (drm_WARN_ON_ONCE(bridge->encoder->dev, + conn_state->connector->hdmi.infoframes.audio.set))
Maybe use drm_err_once() instead to provide the reason for the warning in a string?
+ return -EOPNOTSUPP;
As this check could return error, it should be moved before 'conn_state->hdmi.infoframes.spd.set = false;' to gain a little performance.
+ + /* should not happen, HDR support not enabled */ + if (drm_WARN_ON_ONCE(bridge->encoder->dev, + conn_state->hdmi.infoframes.hdr_drm.set)) + return -EOPNOTSUPP;
I don't think IT6263 chip supports DRM infoframe. The drm_WARN_ON_ONCE() call could make driver readers think that DRM infoframe could be enabled in the future as audio infoframe has the same warning and IT6263 chip does support audio infoframe. So, maybe: /* IT6263 chip doesn't support DRM infoframe. */ conn_state->hdmi.infoframes.hdr_drm.set = false;
+
+ return 0;
+}
+
static const struct drm_bridge_funcs it6263_bridge_funcs = {
+ .atomic_check = it6263_bridge_atomic_check,
.atomic_duplicate_state = drm_atomic_helper_bridge_duplicate_state,
.atomic_destroy_state = drm_atomic_helper_bridge_destroy_state,
.atomic_reset = drm_atomic_helper_bridge_reset,-- Regards, Liu Ying