lists.openwall.net   lists  /  announce  owl-users  owl-dev  john-users  john-dev  passwdqc-users  yescrypt  popa3d-users  /  oss-security  kernel-hardening  musl  sabotage  tlsify  passwords  /  crypt-dev  xvendor  /  Bugtraq  Full-Disclosure  linux-kernel  linux-netdev  linux-ext4  linux-hardening  linux-cve-announce  PHC 
Open Source and information security mailing list archives
 
Hash Suite: Windows password security audit tool. GUI, reports in PDF.
[<prev] [next>] [<thread-prev] [day] [month] [year] [list]
Message-ID: <228c9508-76c3-4772-bf0a-56e85a47af78@collabora.com>
Date: Mon, 23 Sep 2024 14:25:09 +0200
From: AngeloGioacchino Del Regno <angelogioacchino.delregno@...labora.com>
To: Pablo Sun <pablo.sun@...iatek.com>, Rob Herring <robh@...nel.org>,
 Krzysztof Kozlowski <krzk+dt@...nel.org>, Conor Dooley
 <conor+dt@...nel.org>, Matthias Brugger <matthias.bgg@...il.com>,
 Michael Turquette <mturquette@...libre.com>, Stephen Boyd
 <sboyd@...nel.org>, Srinivas Kandagatla <srinivas.kandagatla@...aro.org>
Cc: devicetree@...r.kernel.org, linux-kernel@...r.kernel.org,
 linux-arm-kernel@...ts.infradead.org, linux-mediatek@...ts.infradead.org,
 linux-clk@...r.kernel.org
Subject: Re: [PATCH 5/5] arm64: dts: mediatek: mt8390-genio-700-evk: Enable
 Mali GPU

Il 23/09/24 12:14, Pablo Sun ha scritto:
> Hi Angelo,
> 
> On 9/23/24 16:45, AngeloGioacchino Del Regno wrote:
> [snipped]
>>
>> Is there any real difference between MT8390 and MT8188 in terms of the GPU OPPs?
>>
>> I see that on MT8188, frequencies up to 880MHz want a DVDD_SRAM_GPU of 0.750V,
>> then 0.775/0.762/0.750 (bin1-4/5/6) on 915MHz, and 0.800/0.775/0.750 (bin1-4/5/6)
>> on 950MHz.
>>
>> Those never call for 0.850V...! So is MT8188 (Chromebooks) wrong, or is MT8390
>> different in that?
> 
> To the best of my knowledge, MT8390 and MT8188 has identical GPU
> subsystem. Thus, the OPP table should have no difference.
> 
> To be specific, I list the link to the OPP table of Genio 700 EVK
> (MT8390) in reference in [1]. It should match the setting in Chromebook
> kernel branches.
> 
> The "typical" voltage in the datasheet is the voltage that would work
> for all frequency settings. As long as it is smaller than the maximum
> operating voltage, setting voltages higher than the ones specified in
> the OPP table does not damage the hardware.
> 
> But this 0.85V setting is indeed not optimal. We should follow the
> voltages described in the OPP table, if we want power savings.
> 
> I also considered model the regulator setting with 'regulator-coupled-with' and 
> 'regulator-coupled-max-spread', but I am not entirely sure how to describe the 
> relation that "DVDD_GPU_SRAM should follow DVDD_GPU
> if and only if DVDD_GPU is higher than 0.75V" - should I simply
> set min-voltage to 0.75V and set 'regulator-coupled-with' ?
> 

VSRAM_GPU regulator:

	regulator-min-microvolt = <750000>;
	regulator-max-microvolt = <800000>;
	regulator-coupled-with = <&(vgpu regulator)>;
	regulator-coupled-max-spread = <1000>;

VGPU regulator:
	regulator-min-microvolt = <400000>;
	regulator-max-microvolt = <800000>;
	regulator-coupled-with = <&(vsram_gpu regulator)>;
	regulator-coupled-max-spread = <1000>;

Cheers,
Angelo

> [1]: 
> https://gitlab.com/mediatek/aiot/rity/meta-mediatek-bsp/-/blob/kirkstone/recipes-kernel/dtbo/mt8390/gpu-mali.dts
> 
> Many thanks,
> Pablo
> 




Powered by blists - more mailing lists

Powered by Openwall GNU/*/Linux Powered by OpenVZ