From mboxrd@z Thu Jan 1 00:00:00 1970 From: Claude Code Review Bot To: dri-devel-reviews@example.com Subject: Claude review: drm/msm/dp: Read DPCD and sink count in bridge detect() Date: Mon, 16 Mar 2026 12:12:56 +1000 Message-ID: In-Reply-To: <20260314-hpd-refactor-v5-3-0c8450737d64@oss.qualcomm.com> References: <20260314-hpd-refactor-v5-0-0c8450737d64@oss.qualcomm.com> <20260314-hpd-refactor-v5-3-0c8450737d64@oss.qualcomm.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 Moves `msm_dp_bridge_detect()` from `dp_drm.c` to `dp_display.c` and makes = it actually read DPCD and sink count over AUX rather than just checking `li= nk_ready`. The function is made non-static and declared in `dp_drm.h`. **Issues:** 1. There is a redundant assignment of `dp`: ```c + struct msm_dp_bridge *msm_dp_bridge =3D to_dp_bridge(bridge); + struct msm_dp *dp =3D msm_dp_bridge->msm_dp_display; ... + dp =3D to_dp_bridge(bridge)->msm_dp_display; ``` The second assignment on line 862 duplicates the initialization. Minor, but= should be cleaned up. 2. After the `msm_dp_aux_enable_xfers(priv->aux, true)` call, there is no c= orresponding `msm_dp_aux_enable_xfers(priv->aux, false)` at the `end:` labe= l =E2=80=94 AUX transfers are left enabled after detect returns. This is la= ter addressed in patch 7/9, but it means this patch creates a transient bug= if cherry-picked alone. --- Generated by Claude Code Patch Reviewer