linux-kernel - Re: [PATCH v3 3/4] arm64: dts: apple: t8103: Add i2c nodes

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 for Android: free password hash cracker in your pocket

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

Date:   Wed, 24 Nov 2021 14:42:12 +0900
From:   Hector Martin <marcan@...can.st>
To:     Linus Walleij <linus.walleij@...aro.org>
Cc:     Janne Grunau <j@...nau.net>, Sven Peter <sven@...npeter.dev>,
        Alyssa Rosenzweig <alyssa@...enzweig.io>,
        Rob Herring <robh+dt@...nel.org>,
        linux-arm-kernel@...ts.infradead.org, devicetree@...r.kernel.org,
        linux-kernel@...r.kernel.org
Subject: Re: [PATCH v3 3/4] arm64: dts: apple: t8103: Add i2c nodes

On 24/11/2021 08.26, Linus Walleij wrote:
> On Tue, Nov 23, 2021 at 3:43 PM Hector Martin <marcan@...can.st> wrote:
> 
>> For those following along in the list: the reason why i2c3 was getting
>> stuck is because it seems the unused bus is weakly pulled low on these
>> machines, which jams it.
> 
> That looks like some power saving attempt.
> 
> I suppose that means that even i2c buses that are in use
> could be weakly pulled low when suspending the system
> and maybe even inbetween transactions to save some
> leak current.

Pulled up vs. down dosn't really result in better power savings; neither 
state will necessarily have more leakage. I think it's just that the 
pins are completely disconnected, and there's some very minor leakage to 
ground (megaohms will do) that ends up pulling them down.

Pulling down an I2C bus between transactions is not legal; the idle 
state has to be high.

Apple are actually not very good at configuring GPIOs for power saving; 
e.g. the I/Os for that unused i2c bus still have their input buffers 
turned on, which is a waste of power. If they wanted to save the 
smallest drop of power they'd turn that off. But the effect of this is 
so trivial it probably makes no difference in the context of a laptop, 
nevermind a desktop like the Mac Mini.

-- 
Hector Martin (marcan@...can.st)
Public Key: https://mrcn.st/pub