Re: [External] Re: [PATCH 1/3] thinkpad_acpi: add support for force_discharge

From: Mark Pearson <markpearson@lenovo.com>
To: "Ognjen Galić" <ognjen.galic@outlook.com>,
	"Thomas Koch" <linrunner@gmx.net>,
	"Hans de Goede" <hdegoede@redhat.com>,
	"Nicolo' Piazzalunga" <nicolopiazzalunga@gmail.com>,
	platform-driver-x86@vger.kernel.org,
	"Nitin Joshi1" <njoshi1@lenovo.com>,
	"Sebastian Reichel" <sre@kernel.org>
Cc: <smclt30p@gmail.com>
Subject: Re: [External] Re: [PATCH 1/3] thinkpad_acpi: add support for force_discharge
Date: Mon, 12 Apr 2021 13:10:35 -0400	[thread overview]
Message-ID: <a194b7e1-d812-5c63-373c-db11080d5e1d@lenovo.com> (raw)
In-Reply-To: <VI1PR09MB2302B7C3AD8014CC98D36AA595759@VI1PR09MB2302.eurprd09.prod.outlook.com>


On 07/04/2021 15:01, Ognjen Galić wrote:
> 
> On 07/04/2021 19:48, Mark Pearson wrote:
>> Hi Thomas, Hans and Nicolo
>>
>> On 07/04/2021 08:19, Thomas Koch wrote:
>>> Hi Hans,
>>>
>>>> 1. These features are useful, but not super useful and as such I wonder
>>>> how often they are used and this how well tested the firmware is wrt
>>>> these.
>>>> I have added Mark and Nitin from Lenovo to the Cc. Mark, Nitin, can you
>>>> comment on if it is ok from a firmware pov to try and use the following
>>>> battery related ACPI methods on all thinkpads? :
>>>> #define GET_DISCHARGE    "BDSG"
>>>> #define SET_DISCHARGE    "BDSS"
>>>> #define GET_INHIBIT    "PSSG"
>>>> #define SET_INHIBIT    "BICS"
>>> These ACPI methods are present in (with very few exceptions) all
>>> ThinkPads released since 2012. I am curious to hear what Mark and Nitin
>>> have to say, never read anything official about it.
>> I'm afraid I've not come across these myself before, but will go and ask
>> the firmware team.
>> <For my internal reference LO-1115>
> 
> I received the documents a few years back directly from a Chinese
> contact with all the methods
> and parameters described. I can mail the document to you privately if
> that's needed.
> 
>> It would be good to confirm the implementation details if I can. I found
>> recently that some of the temperature sensors that are read in by
>> thinkpad_acpi from the EC RAM are not temp sensors (patch coming
>> soon....hopefully later today). Hopefully I can check the internal spec
>> and give a thumbs up on the implementation - even if we're not allowed
>> to share the actual paperwork (maybe one day....)
>>
>>> Since 2012 there is also userspace tool tpacpi-bat [1] employing them
>>> along with those for the start/stop threshold.
>>>
>>> My own tool TLP makes use of tpacpi-bat for force_discharge also since
>>> 2012. From my experience in TLP support i can say there's a significant
>>> user base and those who use thresholds also want to use force_discharge
>>> for recalibration from time to time.
>> This probably isn't the right place for the discussion, but I've been
>> meaning to dig into battery management but never really get the time. I
>> know in the windows world that ThinkVantage has extra hooks for setting
>> thresholds and I wanted to see what we can do on the Linux side. If
>> there is anything that would be particularly helpful that is missing
>> please let me know.
>>
>> Thanks
>> Mark
>>
>>>
>>> The patches at hand work flawlessly on my small ThinkPad collection.
>>> [1] https://github.com/teleshoes/tpacpi-bat
>>>
Just as a follow-up - I got some more details on the four ACPI methods
from the firmware team and it all matches up with the code (and indeed
the implementation in tpcacpi-bat).

Mark