Claude review: drm/mediatek: Convert legacy DRM logging to drm_* helpers in mtk_dsi.c

[Claude Code Review Bot <claude-review@example.com>]

Overall Series Review

Subject: drm/mediatek: Convert legacy DRM logging to drm_* helpers in mtk_dsi.c
Author: Abhishek Rajput <abhiraj21put@gmail.com>
Patches: 1
Reviewed: 2026-04-23T10:21:29.034621

---

This patch converts legacy `DRM_INFO()`/`DRM_WARN()`/`DRM_ERROR()` calls in `mtk_dsi.c` to the preferred `drm_info()`/`drm_warn()`/`drm_err()` helpers that take a `struct drm_device *`. The goal is correct and aligns with a longstanding DRM TODO item. However, the patch has two significant bugs and a commit message error.

**Critical issues:**
1. **NULL pointer dereference in `mtk_dsi_host_attach()`**: The patch reads `dsi->bridge.dev` at the top of the function, but at this point the bridge has not been attached to any DRM device yet. `bridge.dev` is only set later during `drm_bridge_attach()` (called indirectly via `component_add()` → `mtk_dsi_bind()` → `mtk_dsi_encoder_init()`). This will be NULL when the error path is hit, causing a crash.

2. **Declaration after statement in `mtk_dsi_switch_to_cmd_mode()`**: The variable declaration `struct drm_device *drm = dsi->bridge.dev;` is placed after two function call statements (`mtk_dsi_irq_data_clear`, `mtk_dsi_set_cmd_mode`). The kernel builds with `-Wdeclaration-after-statement` (and `-Werror` in many configs), so this will cause a build failure. All declarations must come before any statements in a block.

3. **Wrong subsystem in commit message**: The commit message says "This change aligns the radeon driver..." but the patch modifies the MediaTek DSI driver. This is a copy-paste error from a different patch.

---

---
Generated by Claude Code Patch Reviewer