linux-kernel - Re: [PATCH] ACPI / battery: Fix reporting "Not charging" when capacity is 100%

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]

Date:   Mon, 19 Nov 2018 18:12:55 -0800
From:   João Paulo Rechi Vita <jprvita@...il.com>
To:     Pavel Machek <pavel@....cz>
Cc:     Hans de Goede <hdegoede@...hat.com>,
        Daniel Drake <drake@...lessm.com>,
        "Rafael J . Wysocki" <rjw@...ysocki.net>,
        Len Brown <lenb@...nel.org>, linux-acpi@...r.kernel.org,
        Sebastian Reichel <sebastian.reichel@...labora.co.uk>,
        LKML <linux-kernel@...r.kernel.org>, linux@...lessm.com,
        João Paulo Rechi Vita <jprvita@...lessm.com>
Subject: Re: [PATCH] ACPI / battery: Fix reporting "Not charging" when
 capacity is 100%

On Sun, Nov 11, 2018 at 4:22 AM Pavel Machek <pavel@....cz> wrote:
>
> On Sun 2018-11-11 12:57:12, Hans de Goede wrote:
> > Hi,
> >
> > On 11/7/18 5:53 AM, Daniel Drake wrote:
> > >On Mon, Nov 5, 2018 at 1:19 AM Pavel Machek <pavel@....cz> wrote:
> > >>Plus, I don't think "100% charge" is right test for "battery full". At
> > >>least on thinkpads, there's configuration option, and it is common
> > >>_not_ to charge batterry above 95% or so (to increase its lifetime).
> > >
> > >Hans also touched on this area in his response:
> > >
> > >>As for this kernel-side fix I do not believe that fixing thus in
> > >>the kernel is the right thing to do. We try to stay away from
> > >>heuristics using full_charge_capacity in the kernel since that
> > >>is not really reliable / deterministic.
> > >
> > >I'm not fully convinced by this argument though.
> > >
> > >The ACPI spec is not very clear on what conditions you should apply to
> > >decide when the battery is full. Instead, ACPI seems to provide a
> > >pretty decent amount of data, and the decision about whether to
> > >interpret that as "battery full" is left for consumers.
> >
> > Right, but in this case the "discharging" status bit is explicitly
> > set, to me it feels wrong to report "full", when the firmware
> > is reporting "discharging" IMHO, at best we are "not charging"
> > (on AC, below the threshold where a new charge cycle starts) and
> > that is what we are currently reporting.
> >
> > Anu heurstics to decide that "not charging" is close enough to full
> > to report it as full to the user belongs in userspace IMHO.
> >
> > Anyways this ultimately is Rafael's call. If Rafael is ok with this
> > patch then I would like to see Pavel's comment addressed and otherwise
> > it is fine with me.
> >

If we can get to an agreement on this I'll send a v2 without division.

> > Note that we will still often get the case where a laptop is charged,
> > reports full, is unplugged for 5 minutes and then replugged and then
> > reports a capacity of 97% combined with "not charging", so we will
> > still need to fix userspace to handle this.
>

Yes, I agree that should be addressed in userspace, as it is pretty
much a policy decision.

> For the record, I don't think I'm okay with this.
>
> There's nothing special about 100% charge.
>

I don't agree there is nothing special about 100% charge. There is a
separate state to represent battery full for a reason, which is the
user wanting to know when their battery is 100% charged and not being
discharged.

> This changes userland ABI and I don't think it has good enough reasons
to do that.
>

This only changes which state will be reported when the battery is
100% charged and not discharging, it does not introduce / remove any
values. I don't think that is considered ABI change, and on other
hardware like the Dell Latitude 5480,
/sys/class/power_supply/<supply_name>/status already reports "Full"
under these conditions. I still believe it is a bug that makes the ABI
inconsistent across different hardware.

--
João Paulo Rechi Vita
http://about.me/jprvita