[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <010a23e0-4f55-431f-ba34-7b28ff5df43a@rock-chips.com>
Date: Mon, 5 Jan 2026 16:12:00 +0800
From: Chaoyi Chen <chaoyi.chen@...k-chips.com>
To: Nicolas Frattaroli <nicolas.frattaroli@...labora.com>,
Rob Herring <robh@...nel.org>, Krzysztof Kozlowski <krzk+dt@...nel.org>,
Conor Dooley <conor+dt@...nel.org>, Heiko Stuebner <heiko@...ech.de>,
Chaoyi Chen <chaoyi.chen@...k-chips.com>,
Kever Yang <kever.yang@...k-chips.com>,
Sebastian Reichel <sebastian.reichel@...labora.com>,
Cristian Ciocaltea <cristian.ciocaltea@...labora.com>,
Frank Wang <frank.wang@...k-chips.com>, Alexey Charkov <alchark@...il.com>,
Liang Chen <cl@...k-chips.com>, Finley Xiao <finley.xiao@...k-chips.com>,
Elaine Zhang <zhangqing@...k-chips.com>,
Yifeng Zhao <yifeng.zhao@...k-chips.com>
Cc: devicetree@...r.kernel.org, linux-arm-kernel@...ts.infradead.org,
linux-rockchip@...ts.infradead.org, linux-kernel@...r.kernel.org
Subject: Re: [PATCH] arm64: dts: rockchip: Fix wrong register range of rk3576
gpu
Hello Nicolas,
On 1/5/2026 3:55 PM, Nicolas Frattaroli wrote:
> On Tuesday, 30 December 2025 10:02:46 Central European Standard Time Chaoyi Chen wrote:
>> From: Chaoyi Chen <chaoyi.chen@...k-chips.com>
>>
>> According to RK3576 TRM part1 page13, the size of the GPU registers
>> is 128 KB.
>>
>> Fixes: 57b1ce903966 ("arm64: dts: rockchip: Add rk3576 SoC base DT")
>> Signed-off-by: Chaoyi Chen <chaoyi.chen@...k-chips.com>
>> ---
>> arch/arm64/boot/dts/rockchip/rk3576.dtsi | 2 +-
>> 1 file changed, 1 insertion(+), 1 deletion(-)
>>
>> diff --git a/arch/arm64/boot/dts/rockchip/rk3576.dtsi b/arch/arm64/boot/dts/rockchip/rk3576.dtsi
>> index 6284e7bdc442..b375015f0662 100644
>> --- a/arch/arm64/boot/dts/rockchip/rk3576.dtsi
>> +++ b/arch/arm64/boot/dts/rockchip/rk3576.dtsi
>> @@ -1271,7 +1271,7 @@ power-domain@...576_PD_VO1 {
>>
>> gpu: gpu@...00000 {
>> compatible = "rockchip,rk3576-mali", "arm,mali-bifrost";
>> - reg = <0x0 0x27800000 0x0 0x200000>;
>> + reg = <0x0 0x27800000 0x0 0x20000>;
>> assigned-clocks = <&scmi_clk SCMI_CLK_GPU>;
>> assigned-clock-rates = <198000000>;
>> clocks = <&cru CLK_GPU>;
>>
>
> This is only true if you only consider the GPU_CONTROL and
> JOB_CONTROL register ranges, and leave out the MMU_STAGE1
> and MMU_STAGE2 ranges. I don't know if those need to be
> mapped, since the MMU control registers are < 0x2000.
>
> What do other bifrost devices do?
>
I don't quite understand your point. And this size is 0x20000,
not 0x2000. Before the change it was 0x200000, which overlaps the
address space of some hardware that comes after it like JPEG.
Could you explain it again? Thank you.
> Kind regards,
> Nicolas Frattaroli
>
>
>
--
Best,
Chaoyi
--
Best,
Chaoyi
Powered by blists - more mailing lists