This series adds support for Novatek NT51021-based panels 
for upcoming xiaomi-clover submission.

Although according to datasheet, the IС has three power rails 
in the use case of xiaomi-clover, all of them are wired to one 
fixed regulator.

Maybe I should create a dt option for enabling CABC for panel.

Signed-off-by: Nickolay Goppen <setotau@mainlining.org>
---
Changes in v3:
- Fixed bindings (Krzysztof)
- Fixed probe function (Sashiko)
- Added BOE-TV101WUM-NM0 support (Ryan)
- Link to v2: https://patch.msgid.link/20260911-qcom-novatek-nt51021-panels-v2-0-135e3190a7fb@mainlining.org

Changes in v2:
- Use devm_functions and renamed structs and functions (Neil)
- Added delay before deasserting reset for stable initialization (Sashiko and Ryan)
- Fixed dt-bindings (Sashiko and Neil)
- Added a comment about set_display_on command
- Link to v1: https://patch.msgid.link/20260829-qcom-novatek-nt51021-panels-v1-0-cc1ca61bd261@mainlining.org

To: Nickolay Goppen <setotau@mainlining.org>
To: Neil Armstrong <neil.armstrong@linaro.org>
To: Jessica Zhang <jesszhan0024@gmail.com>
To: David Airlie <airlied@gmail.com>
To: Simona Vetter <simona@ffwll.ch>
To: Maarten Lankhorst <maarten.lankhorst@linux.intel.com>
To: Maxime Ripard <mripard@kernel.org>
To: Thomas Zimmermann <tzimmermann@suse.de>
To: Rob Herring <robh@kernel.org>
To: Krzysztof Kozlowski <krzk+dt@kernel.org>
To: Conor Dooley <conor+dt@kernel.org>
To: Matthias Brugger <matthias.bgg@gmail.com>
To: AngeloGioacchino Del Regno <angelogioacchino.delregno@collabora.com>
Cc: dri-devel@lists.freedesktop.org
Cc: devicetree@vger.kernel.org
Cc: linux-kernel@vger.kernel.org
Cc: Ryan Brue <ryanbrue.dev@gmail.com>
Cc: Alexey Minnekhanov <alexeymin@minlexx.ru>
Cc: linux-arm-kernel@lists.infradead.org
Cc: linux-mediatek@lists.infradead.org

---
Nickolay Goppen (3):
      dt-bindings: display: panel: Add Novatek NT51021
      drivers: gpu: drm: panel: Add BOE NT51021 driver
      MAINTAINERS: Add myself and Ryan Brue as maintainers of NT51021 panel driver

 .../bindings/display/panel/novatek,nt51021.yaml    |  87 +++
 MAINTAINERS                                        |   7 +
 drivers/gpu/drm/panel/Kconfig                      |   9 +
 drivers/gpu/drm/panel/Makefile                     |   1 +
 drivers/gpu/drm/panel/panel-novatek-nt51021.c      | 855 +++++++++++++++++++++
 5 files changed, 959 insertions(+)
---
base-commit: 99cc957e3ef3cccd72d2119c8eacbab39af0d1d6
change-id: 20260829-qcom-novatek-nt51021-panels-e2f9db85a967

Best regards,
--  
Nickolay Goppen <setotau@mainlining.org>

Tested-by: Ryan Brue <ryanbrue.dev@gmail.com> # Amazon Fire HD 10 
(2017), boe,tv101wum-nm0

On the CABC teardown for the nm0 variant that Sashiko found: the write 
is a single DSI generic command to the panel IC just before it is 
powered off, so there is nothing on I2C for it to collide with (the 
LP8557 backlight is a different device and is already off by then). On 
the Fire HD 10 the full disable/unprepare then prepare/enable path runs 
on every DPMS cycle and system suspend, and the panel comes back cleanly 
every time. Skipping the call for variants that never turn CABC on would 
be tidier, though, and I am fine either way; I'll let it be your call 
Nickolay. Thanks for pulling in my patches!

On Sun, Sep 13, 2026 at 12:23:46PM -0500, Ryan Brue wrote:
> Tested-by: Ryan Brue <ryanbrue.dev@gmail.com> # Amazon Fire HD 10 (2017),
> boe,tv101wum-nm0

This tag is applicable only to one patch out of tree, so just respond
there instead of causing scripts to get this tag to all patches.

Best regards,
Krzysztof

On 9/15/26 5:39 AM, Krzysztof Kozlowski wrote:

> On Sun, Sep 13, 2026 at 12:23:46PM -0500, Ryan Brue wrote:
>> Tested-by: Ryan Brue <ryanbrue.dev@gmail.com> # Amazon Fire HD 10 (2017),
>> boe,tv101wum-nm0
> This tag is applicable only to one patch out of tree, so just respond
> there instead of causing scripts to get this tag to all patches.
>
> Best regards,
> Krzysztof
My apologies, I'm new to kernel development, and didn't realize, but in 
hindsight it makes sense. Thanks for being kind about it, it won't 
happen again.

Best regards,
Ryan