linux-kernel - Re: [PATCH 2/2] rtc: pcf2127: add battery-low INTn generation

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: <20190403161529.GS22216@piout.net>
Date:   Wed, 3 Apr 2019 18:15:29 +0200
From:   Alexandre Belloni <alexandre.belloni@...tlin.com>
To:     Flavio Suligoi <f.suligoi@...m.it>
Cc:     Alessandro Zummo <a.zummo@...ertech.it>,
        Rob Herring <robh+dt@...nel.org>,
        Mark Rutland <mark.rutland@....com>,
        "linux-rtc@...r.kernel.org" <linux-rtc@...r.kernel.org>,
        "devicetree@...r.kernel.org" <devicetree@...r.kernel.org>,
        "linux-kernel@...r.kernel.org" <linux-kernel@...r.kernel.org>
Subject: Re: [PATCH 2/2] rtc: pcf2127: add battery-low INTn generation

On 03/04/2019 16:09:14+0000, Flavio Suligoi wrote:
> Hi,
> 
> > On 03/04/2019 15:49:03+0000, Flavio Suligoi wrote:
> > > > Then, you should probably not enable BLIE because this will cause
> > issues
> > > > with the alarm functionnality.. It is certainly enough to use
> > > > RTC_VL_READ periodically.
> > >
> > > We use the nINT signaling solution because of this pin, in addition
> > > to be used by the CPU, can be also connected to an external connector,
> > > available for the final user.
> > > Anyway, even if the BLIE is set, the sw low voltage alarm works,
> > > with the message (displayed about every 12 minutes):
> > >
> > 
> > I agree the DT property makes sense when the nINT pin is not connected
> > to the CPU. But if it is, then you have an issue that nINT will be
> > pulled low until the user changes the battery, meaning that you will
> > not get any alarm interrupt anymore, possibly leading to a system that
> > is not waking up anymore.
> 
> Ah, ok, thanks for the info. 
> I know this, but in our specific case, this is not a problem,
> since we don't use the nINT for other purposes, but only for
> a battery low indicator. On the contrary, in our case, it's better
> that the alarm signal remains low until the battery is changed.
> 
> Anyway, I can specify this collateral effect in a specific file for the
> Pcf2127 in the DT bindings Documentation. What do you think?
> 

This is fine as-is, I'll handle that in my series adding alarm support
because there will be no issues until then.

-- 
Alexandre Belloni, Bootlin
Embedded Linux and Kernel engineering
https://bootlin.com