linux-kernel - Re: [PATCH v2 4/6] ARM: at91/dt: sama5d3: add i2c gpio pinctrl

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: <e22772f8-9e6d-002d-98d7-414136a32439@axentia.se>
Date:   Sat, 4 Jan 2020 22:39:48 +0000
From:   Peter Rosin <peda@...ntia.se>
To:     "Codrin.Ciubotariu@...rochip.com" <Codrin.Ciubotariu@...rochip.com>,
        "linux-i2c@...r.kernel.org" <linux-i2c@...r.kernel.org>,
        "devicetree@...r.kernel.org" <devicetree@...r.kernel.org>,
        "linux-arm-kernel@...ts.infradead.org" 
        <linux-arm-kernel@...ts.infradead.org>,
        "linux-kernel@...r.kernel.org" <linux-kernel@...r.kernel.org>
CC:     "kamel.bouhara@...tlin.com" <kamel.bouhara@...tlin.com>,
        "wsa@...-dreams.de" <wsa@...-dreams.de>,
        "Nicolas.Ferre@...rochip.com" <Nicolas.Ferre@...rochip.com>,
        "alexandre.belloni@...tlin.com" <alexandre.belloni@...tlin.com>,
        "Ludovic.Desroches@...rochip.com" <Ludovic.Desroches@...rochip.com>,
        "robh@...nel.org" <robh@...nel.org>
Subject: Re: [PATCH v2 4/6] ARM: at91/dt: sama5d3: add i2c gpio pinctrl

On 2020-01-03 10:49, Codrin.Ciubotariu@...rochip.com wrote:
> From: Kamel Bouhara <kamel.bouhara@...tlin.com>
> 
> Add the i2c gpio pinctrls to support the i2c bus recovery
> 
> Signed-off-by: Kamel Bouhara <kamel.bouhara@...tlin.com>
> ---
> 
> Changes in v2:
>  - none;
> 
>  arch/arm/boot/dts/sama5d3.dtsi | 33 ++++++++++++++++++++++++++++++---
>  1 file changed, 30 insertions(+), 3 deletions(-)
> 

*snip*

> @@ -639,6 +648,12 @@
>  							<AT91_PIOA 30 AT91_PERIPH_A AT91_PINCTRL_NONE	/* PA30 periph A TWD0 pin, conflicts with URXD1, ISI_VSYNC */
>  							 AT91_PIOA 31 AT91_PERIPH_A AT91_PINCTRL_NONE>;	/* PA31 periph A TWCK0 pin, conflicts with UTXD1, ISI_HSYNC */
>  					};
> +
> +					pinctrl_i2c0_gpio: i2c0-gpio {
> +						atmel,pins =
> +							<AT91_PIOA 30 AT91_PERIPH_GPIO AT91_PINCTRL_PULL_UP
> +							 AT91_PIOA 31 AT91_PERIPH_GPIO AT91_PINCTRL_PULL_UP>;
> +					};

I'm curious, but why are pull-ups suddenly needed just because the pins are
used for GPIO recovery? Why are pull-ups not needed when the pins are used
by the I2C peripheral device(s)?

Given figure 27-2 "I/O Line Control Logic" in my SAMA5D3 datasheet, I see
no difference as to how and why the pull-ups are applied depending on what
the current function of the pin is. So, if the I2C bus works w/o pulls, bus
recovery using GPIO must also work w/o pulls.

I.e. the device tree requires you to have external pull-ups on the I2C bus
anyway, so why bother with internal pull-ups for the recovery case?

Changing pull-up settings just for recovery feels like something that will
inevitably create hard to debug surprises at the least opportune time...

Or am I missing something?

(I'm focusing on SAMA5D3 since that is what I happen to work with,
 but the same question appears to apply for SAMA5D2 and SAMA5D4...)

Cheers,
Peter