From mboxrd@z Thu Jan  1 00:00:00 1970
From: Claude Code Review Bot <claude-review@example.com>
To: dri-devel-reviews@example.com
Subject: Claude review: drm/bridge: dw-hdmi: Replace calls to deprecated HDMI/audio detection functions
Date: Fri, 27 Feb 2026 12:14:29 +1000
Message-ID: <review-patch1-20260226120626.108378-1-karanja99erick@gmail.com>
In-Reply-To: <20260226120626.108378-1-karanja99erick@gmail.com>
References: <20260226120626.108378-1-karanja99erick@gmail.com> <20260226120626.108378-1-karanja99erick@gmail.com>
X-Mailer: Claude Code Patch Reviewer
Content-Type: text/plain; charset="utf-8"
Content-Transfer-Encoding: quoted-printable
MIME-Version: 1.0

Patch Review

**Subject line is misleading.** The commit message says "Replace calls to d=
eprecated...functions", but the patch only *removes* the old calls =E2=80=
=94 it does not replace anything. The consumers are left reading stale (alw=
ays-false) fields.

**Critical: Remaining consumers not updated.** After this patch, the follow=
ing code still references the now-never-assigned fields:

1. `dw-hdmi.c:2061` =E2=80=94 `hdmi_av_composer()`:
   ```c
   inv_val |=3D hdmi->sink_is_hdmi ?
       HDMI_FC_INVIDCONF_DVI_MODEZ_HDMI_MODE :
       HDMI_FC_INVIDCONF_DVI_MODEZ_DVI_MODE;
   ```
   Will always select DVI mode.

2. `dw-hdmi.c:2294` =E2=80=94 `dw_hdmi_setup()`:
   ```c
   hdmi->hdmi_data.rgb_limited_range =3D hdmi->sink_is_hdmi &&
       drm_default_rgb_quant_range(mode) =3D=3D
       HDMI_QUANTIZATION_RANGE_LIMITED;
   ```
   Will always set `rgb_limited_range =3D false`.

3. `dw-hdmi.c:2316` =E2=80=94 `dw_hdmi_setup()`:
   ```c
   if (hdmi->sink_has_audio) {
   ```
   Audio will never be configured.

4. `dw-hdmi.c:2325` =E2=80=94 `dw_hdmi_setup()`:
   ```c
   if (hdmi->sink_is_hdmi) {
   ```
   AVI InfoFrame, vendor-specific infoframe, and DRM infoframe configuratio=
n will be skipped entirely.

**What a correct patch would need to do:**
- Remove the `sink_is_hdmi` and `sink_has_audio` fields from `struct dw_hdm=
i` (lines 161-162).
- Change all four consumer sites to read from `connector->display_info.is_h=
dmi` and `connector->display_info.has_audio` instead. Note that `dw_hdmi_se=
tup()` and `hdmi_av_composer()` already receive a `const struct drm_connect=
or *connector` parameter (or `const struct drm_display_info *display`), so =
the information is accessible =E2=80=94 it just needs to be wired up.

**Additional note on the bridge path:** When dw-hdmi operates via `dw_hdmi_=
bridge_edid_read()` (line 2990), the bridge framework calls `drm_edid_conne=
ctor_update()` on behalf of the caller, so the approach is sound =E2=80=94 =
but only if the consumers are updated.

**Minor:** The removed `dev_dbg` logging of EDID width/height (useful for d=
ebug) could be kept or moved, but this is cosmetic and not a blocker.

---
Generated by Claude Code Patch Reviewer