linux-kernel - Re: [PATCH] power: supply: qcom_battmgr: remove bogus do

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: <c65e2dbc-c8f9-4481-85f1-4d1eb140a05f@mercury.local>
Date:   Sat, 4 Mar 2023 02:37:03 +0100
From:   Sebastian Reichel <sre@...nel.org>
To:     Linus Torvalds <torvalds@...ux-foundation.org>
Cc:     Nathan Chancellor <nathan@...nel.org>,
        Andy Gross <agross@...nel.org>,
        Bjorn Andersson <andersson@...nel.org>,
        Arnd Bergmann <arnd@...db.de>,
        Konrad Dybcio <konrad.dybcio@...aro.org>,
        Neil Armstrong <neil.armstrong@...aro.org>,
        linux-arm-msm@...r.kernel.org, linux-pm@...r.kernel.org,
        linux-kernel@...r.kernel.org
Subject: Re: [PATCH] power: supply: qcom_battmgr: remove bogus do_div()

Hi,

On Wed, Mar 01, 2023 at 10:50:33AM -0800, Linus Torvalds wrote:
> On Wed, Mar 1, 2023 at 10:16 AM Nathan Chancellor <nathan@...nel.org> wrote:
> > Would you be able to take this patch directly? It seems obviously
> > correctTM, has an ack from Sebastian [1], and without it, 32-bit
> > allmodconfig builds are broken [2] (the other warning in that log has a
> > fix on the way to you soon).
> 
> Ok, I've taken it directly.
> 
> However, the whole "seems obviously correct" is true in the sense that
> it now doesn't complain (and doesn't overwrite an "int" with a 64-bit
> value.
> 
> The actual code still looks odd. Is that return value for
> BATT_CAPACITY truly in that odd "hundredths of percent" format,
> where dividing it by 100 gives you whole percent?
> 
> Because "hundredths of percent" strikes me as a very odd interface.
> Even for some firmware interface. I realize that anything is possible
> with strange firmware, but still...

I don't have the documentation for this Qualcomm interface, but I
remember somebody from Qualcomm asking for a power-supply property
exposing "hundredths of percent" to userspace some years ago (with
the rationale, that percent was not precise enough).

For reference: The upstream solution for that is exposing ENERGY_NOW
and ENERGY_FULL in µWh (or CHARGE_NOW + CHARGE_FULL in µAh depending
on the fuel-gauge's capabilities), which is even more precise.

-- Sebastian

Download attachment "signature.asc" of type "application/pgp-signature" (834 bytes)