Redmi 4A (rolex) is like Redmi 5A with small differences like charging,
fuel gauge, different speaker codec configuration and display.

Signed-off-by: Barnabás Czémán <barnabas.czeman@mainlining.org>
---
 arch/arm64/boot/dts/qcom/Makefile                 |  1 +
 arch/arm64/boot/dts/qcom/msm8917-xiaomi-rolex.dts | 15 +++++++++++++++
 2 files changed, 16 insertions(+)

diff --git a/arch/arm64/boot/dts/qcom/Makefile b/arch/arm64/boot/dts/qcom/Makefile
index 6d87be639aac..20e161e843ed 100644
--- a/arch/arm64/boot/dts/qcom/Makefile
+++ b/arch/arm64/boot/dts/qcom/Makefile
@@ -82,6 +82,7 @@ dtb-$(CONFIG_ARCH_QCOM)	+= msm8916-wingtech-wt86528.dtb
 dtb-$(CONFIG_ARCH_QCOM)	+= msm8916-wingtech-wt88047.dtb
 dtb-$(CONFIG_ARCH_QCOM)	+= msm8916-yiming-uz801v3.dtb
 dtb-$(CONFIG_ARCH_QCOM)	+= msm8917-xiaomi-riva.dtb
+dtb-$(CONFIG_ARCH_QCOM)	+= msm8917-xiaomi-rolex.dtb
 dtb-$(CONFIG_ARCH_QCOM)	+= msm8929-wingtech-wt82918hd.dtb
 dtb-$(CONFIG_ARCH_QCOM)	+= msm8937-xiaomi-land.dtb
 dtb-$(CONFIG_ARCH_QCOM)	+= msm8939-asus-z00t.dtb
diff --git a/arch/arm64/boot/dts/qcom/msm8917-xiaomi-rolex.dts b/arch/arm64/boot/dts/qcom/msm8917-xiaomi-rolex.dts
new file mode 100644
index 000000000000..f0b72d9878c4
--- /dev/null
+++ b/arch/arm64/boot/dts/qcom/msm8917-xiaomi-rolex.dts
@@ -0,0 +1,15 @@
+// SPDX-License-Identifier: GPL-2.0-only
+/*
+ * Copyright (c) 2026, Barnabas Czeman
+ */
+
+/dts-v1/;
+
+#include "msm8917-xiaomi-wingtech.dtsi"
+
+/ {
+	model = "Xiaomi Redmi 4A (rolex)";
+	compatible = "xiaomi,rolex", "qcom,msm8917";
+
+	qcom,board-id = <0x1000b 1>;
+};

-- 
2.53.0

On 3/10/26 6:21 PM, Barnabás Czémán wrote:
> Redmi 4A (rolex) is like Redmi 5A with small differences like charging,
> fuel gauge, different speaker codec configuration and display.
> 
> Signed-off-by: Barnabás Czémán <barnabas.czeman@mainlining.org>
> ---

I see the Go has some differences in regulator configuration.

Does the 4A's one match the existing 5A?

Konrad

On 2026-03-11 10:16, Konrad Dybcio wrote:
> On 3/10/26 6:21 PM, Barnabás Czémán wrote:
>> Redmi 4A (rolex) is like Redmi 5A with small differences like 
>> charging,
>> fuel gauge, different speaker codec configuration and display.
>> 
>> Signed-off-by: Barnabás Czémán <barnabas.czeman@mainlining.org>
>> ---
> 
> I see the Go has some differences in regulator configuration.
> 
> Does the 4A's one match the existing 5A?
Yes they are same, l22 is vana l23 is vdig for camera sensor.
riva and rolex is using 2.8v for vana and 1.2v for vdig.
tiare is using 2.85v for vana and 1.2 v for vdig so l22 min-voltage also 
should be increased
but l23 max-voltage increase is unnecessary, there is no reason to keep 
downstream value there.

Here is some proof 
https://github.com/Mi-Thorium/kernel_devicetree_xiaomi-msm8937/blob/dts/mi8937/4.9/master/wingtech/msm8917/tiare/camera.dtsi#L16C1-L23C65
> 
> Konrad

On 3/11/26 3:38 PM, barnabas.czeman@mainlining.org wrote:
> On 2026-03-11 10:16, Konrad Dybcio wrote:
>> On 3/10/26 6:21 PM, Barnabás Czémán wrote:
>>> Redmi 4A (rolex) is like Redmi 5A with small differences like charging,
>>> fuel gauge, different speaker codec configuration and display.
>>>
>>> Signed-off-by: Barnabás Czémán <barnabas.czeman@mainlining.org>
>>> ---
>>
>> I see the Go has some differences in regulator configuration.
>>
>> Does the 4A's one match the existing 5A?
> Yes they are same, l22 is vana l23 is vdig for camera sensor.
> riva and rolex is using 2.8v for vana and 1.2v for vdig.
> tiare is using 2.85v for vana and 1.2 v for vdig so l22 min-voltage also should be increased
> but l23 max-voltage increase is unnecessary, there is no reason to keep downstream value there.
> 
> Here is some proof https://github.com/Mi-Thorium/kernel_devicetree_xiaomi-msm8937/blob/dts/mi8937/4.9/master/wingtech/msm8917/tiare/camera.dtsi#L16C1-L23C65

FWIW it's easiest to work with decompiled DTs, as the include chains are
often very complex. You can then throw them into ./scripts/dtx_diff or
similar

Konrad