linux-kernel - Re: [PATCH v2 5/8] power: supply: qcom_battmgr: Add charge control support

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] [thread-next>] [day] [month] [year] [list]

Message-ID: <bc1daec6-9198-43af-b5a6-c1057d7f2392@oss.qualcomm.com>
Date: Tue, 10 Jun 2025 14:19:58 +0200
From: Konrad Dybcio <konrad.dybcio@....qualcomm.com>
To: Dmitry Baryshkov <dmitry.baryshkov@....qualcomm.com>,
        Fenglin Wu <fenglin.wu@....qualcomm.com>,
        Konrad Dybcio <konrad.dybcio@....qualcomm.com>
Cc: György Kurucz <me@...uczgy.com>,
        Sebastian Reichel <sre@...nel.org>,
        Bjorn Andersson <andersson@...nel.org>,
        Konrad Dybcio <konradybcio@...nel.org>, Rob Herring <robh@...nel.org>,
        Krzysztof Kozlowski <krzk+dt@...nel.org>,
        Conor Dooley
 <conor+dt@...nel.org>,
        Heikki Krogerus <heikki.krogerus@...ux.intel.com>,
        Greg Kroah-Hartman <gregkh@...uxfoundation.org>,
        Subbaraman Narayanamurthy <subbaraman.narayanamurthy@....qualcomm.com>,
        David Collins <david.collins@....qualcomm.com>,
        linux-pm@...r.kernel.org, linux-kernel@...r.kernel.org,
        linux-arm-msm@...r.kernel.org, kernel@....qualcomm.com,
        devicetree@...r.kernel.org, linux-usb@...r.kernel.org
Subject: Re: [PATCH v2 5/8] power: supply: qcom_battmgr: Add charge control
 support

On 6/9/25 9:17 AM, Dmitry Baryshkov wrote:
> On 09/06/2025 05:39, Fenglin Wu wrote:
>>
>> On 6/7/2025 5:46 PM, Konrad Dybcio wrote:
>>> On 6/3/25 12:37 PM, Dmitry Baryshkov wrote:
>>>> On Tue, Jun 03, 2025 at 01:48:11PM +0800, Fenglin Wu wrote:
>>>>> On 5/31/2025 6:36 PM, György Kurucz wrote:
>>>>>>> Add charge control support for SM8550 and X1E80100.
>>>>>> Thank you for this, tested on my Lenovo Yoga Slim 7x, the limiting works
>>>>>> well, I finally don't have to worry about leaving my laptop plugged in
>>>>>> for too long.
>>>>>>
>>>>>> One small thing I noticed is that after setting the sysfs values and
>>>>>> rebooting, they report 0 again. The limiting appears to stay in effect
>>>>>> though, so it seems that the firmware does keep the values, but Linux
>>>>>> does not read them back. Indeed, looking at the code, it seems that
>>>>>> actually reading back the values is only implemented for the SM8550.
>>>>> Right.
>>>>>
>>>>> Based on offline information, X1E80100 doesn't support reading back those
>>>>> threshold values in battery management firmware, so I can only use the
>>>>> cached values for sysfs read.
>>>> Which limits usablity of the attribute, it is now impossible to identify
>>>> whether it is enabled or disabled. Is there a chance of fixing that for
>>>> the X1E80100 platform?
>>> Is there a chance we store that value in SDAM and can read it back?
>>>
>>> Konrad
>>
>> The thresholds are stored in PMIC SDAM registers by ADSP after receiving the set requests, and ADSP reads them back during initialization. This is why ADSP retains them upon device reboot.
>>
>> I spoke with the battery management firmware team, and they have no plans to update the battery management firmware for X1E80100 further. Consequently, they cannot provide any interfaces to read these thresholds through PMIC Glink.
>>
>> Reading them from the existing SDAM registers requires adding "nvmem- cells" DT properties to specify the SDAM registers. However, the "pmic_glink.power-supply" device is an auxiliary device created by the pmic_glink driver and does not have an associated DT node. Is there any method to create a DT node and add DT properties for an auxiliary device?
> 
> Auxiliary-bus devices don't have their OF nodes. Instead they use the main device's of node thanks to the call to device_set_of_node_from_dev().

i.e. something like this is what we want:

---- socname.dtsi ----

pmic-glink {
	compatible = ...;
	[...]

	nvmem-cells = <&charge_limit_lower>, <&charge_limit_upper>;
	nvmem-cell-names = "charge-limit-lower", "charge-limit-upper";
}

---------------------

if you have better names for these cells than what i put here as
placeholders, go for it - just make sure to also add the -names
counterpart, as we won't be able to rely on just indices in the long run

Konrad