linux-kernel - Re: [PATCH 1/2] gpio: add pin biasing and drive mode to gpiolib

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]

Message-ID: <BANLkTin_e1Kkn=KGvnqCHvnJVZczRwdnVA@mail.gmail.com>
Date:	Thu, 21 Apr 2011 08:48:31 +0200
From:	Linus Walleij <linus.walleij@...aro.org>
To:	Ben Nizette <bn@...sdigital.com>
Cc:	Alan Cox <alan@...rguk.ukuu.org.uk>, linux-kernel@...r.kernel.org,
	Grant Likely <grant.likely@...retlab.ca>,
	Lee Jones <lee.jones@...aro.org>,
	linux-arm-kernel@...ts.infradead.org
Subject: Re: [PATCH 1/2] gpio: add pin biasing and drive mode to gpiolib

2011/4/21 Ben Nizette <bn@...sdigital.com>:

>> This patch is about biasing and drive modes. We need to alter
>> these at runtime (from board code, indeed) due to the fact that
>> when you go to sleep e.g. floating a pin yeilds better power
>> characteristics.
>
> This is actually an interesting case because floating pins yeild
> /worse/ power characteristics (each transistor of the push-pull
> is on a little bit and you get a path straight through) [1].  To
> get good power performance you want to pull an input pin high or
> low but which of those two directions depends on external
> constraints, i.e. the board.

Yeah, mea culpa, I twisted it around in my head. I mean the
reverse. So this is why there are GPIO_CONFIG_BIAS_HIGH
and GPIO_CONFIG_BIAS_GROUND in the predefined pin
bias states.

(I do get it, don't worry, I was in electroscience before I
turned computer science actually...)

>  This is a case where the driver
> should /not/ go playing with things it can't fully understand.
>
> Perhaps one of the properties that a board can set in a gpio chip
> driver is the suspend state and have suspend/resume hooks in the
> gpio chip take care of setting things up on each side.

Yes, the API is not only for drivers, it's for boards alright.

Pushing it to drivers/gpio give us two advantages:

1. Depopulate the overpopulated arch/arm/* etc trees
2. Give an overview so the GPIO maintainer and others can
  think about consolidations
3. Clear cut drivers that can have runtime PM hooks etc...

Linus Walleij
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/