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/panel: himax-hx83102: restore MODE_LPM after sending disable cmds
Date: Tue, 28 Apr 2026 15:33:35 +1000
Message-ID: <review-patch1-20260425165751.1716569-1-zhengxingda@iscas.ac.cn>
In-Reply-To: <20260425165751.1716569-1-zhengxingda@iscas.ac.cn>
References: <20260425165751.1716569-1-zhengxingda@iscas.ac.cn> <20260425165751.1716569-1-zhengxingda@iscas.ac.cn>
X-Mailer: Claude Code Patch Reviewer
Content-Type: text/plain; charset="utf-8"
Content-Transfer-Encoding: quoted-printable
MIME-Version: 1.0

Patch Review

**Analysis of the bug:**

At probe time (line 1173-1174), `mode_flags` is initialized with `MIPI_DSI_=
MODE_LPM` set:
```c
dsi->mode_flags =3D MIPI_DSI_MODE_VIDEO | MIPI_DSI_MODE_VIDEO_SYNC_PULSE |
                  MIPI_DSI_MODE_LPM;
```

In `hx83102_disable()`, `MIPI_DSI_MODE_LPM` is cleared at line 935:
```c
dsi->mode_flags &=3D ~MIPI_DSI_MODE_LPM;
```

Before this patch, the flag was never restored. The `unprepare` function (l=
ine 947) doesn't issue any DSI commands =E2=80=94 it only toggles GPIOs and=
 regulators =E2=80=94 so the missing flag persists into the next `prepare`/=
`init` cycle. Since the init sequences send many DSI commands expecting LP =
mode, this breaks the second panel enable.

**The fix** (line 940):
```c
dsi->mode_flags |=3D MIPI_DSI_MODE_LPM;
```

This is placed after the two DSI DCS commands (`set_display_off` and `enter=
_sleep_mode`) and before the `mipi_dsi_msleep()`. The placement is correct =
=E2=80=94 the disable commands that need HS mode have already been sent, an=
d restoring LPM before the sleep is harmless (the sleep helper only touches=
 the error accumulator and calls `msleep`, it doesn't send a DSI command th=
at would be affected by the mode flag).

**Consistency with existing code:** The backlight functions in the same fil=
e (`hx83102_bl_update_status` at lines 1064-1070 and `hx83102_bl_get_bright=
ness` at lines 1081-1087) use the exact same clear-before/restore-after pat=
tern. This patch makes `hx83102_disable` consistent with those.

**Minor observation:** The commit message and `Fixes:` tag are well-formed.=
 The `Cc: stable@vger.kernel.org # 6.11+` annotation is appropriate since t=
he bug was introduced by the refactoring in commit `0ef94554dc40`.

**No issues found.**

Reviewed-by tag recommendation: This patch is ready to be applied as-is.

---
Generated by Claude Code Patch Reviewer