linux-kernel - Re: [PATCH 1/6] ASoC: wcd938x: switch to using gpiod API

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: <d746ee5f-283d-44ce-b72c-18c8955d38b1@sirena.org.uk>
Date:   Thu, 20 Apr 2023 14:00:06 +0100
From:   Mark Brown <broonie@...nel.org>
To:     Krzysztof Kozlowski <krzysztof.kozlowski@...aro.org>
Cc:     Vinod Koul <vkoul@...nel.org>,
        Bard Liao <yung-chuan.liao@...ux.intel.com>,
        Pierre-Louis Bossart <pierre-louis.bossart@...ux.intel.com>,
        Sanyog Kale <sanyog.r.kale@...el.com>,
        Andy Gross <agross@...nel.org>,
        Bjorn Andersson <andersson@...nel.org>,
        Konrad Dybcio <konrad.dybcio@...aro.org>,
        Liam Girdwood <lgirdwood@...il.com>,
        Jaroslav Kysela <perex@...ex.cz>,
        Takashi Iwai <tiwai@...e.com>, alsa-devel@...a-project.org,
        linux-kernel@...r.kernel.org, linux-arm-msm@...r.kernel.org,
        Dmitry Torokhov <dmitry.torokhov@...il.com>,
        Patrick Lai <quic_plai@...cinc.com>
Subject: Re: [PATCH 1/6] ASoC: wcd938x: switch to using gpiod API

On Thu, Apr 20, 2023 at 02:30:17PM +0200, Krzysztof Kozlowski wrote:
> On 20/04/2023 13:58, Mark Brown wrote:
> > On Thu, Apr 20, 2023 at 12:16:12PM +0200, Krzysztof Kozlowski wrote:

> >> -	gpio_direction_output(wcd938x->reset_gpio, 0);
> >> -	/* 20us sleep required after pulling the reset gpio to LOW */
> >> +	gpiod_set_value_cansleep(wcd938x->reset_gpio, 1);
> >> +	/* 20us sleep required after asserting the reset gpio */

> > This is inverting the sense of the GPIO in the API from active low to
> > active high which will mean we're introducing a new reliance on having
> > the signal described as active low in DT.  That's an ABI concern.

> It's bringing it to the correct level. Old code was not respecting the
> DTS thus if such DTS came with inverted design, the driver would not work.

Sure, but OTOH if the user didn't bother specifying as active low it
would work.  I suspect it's more likely that someone missed a flag that
had no practical impact in DT than that someone would add an inverter to
their design.

> We were already fixing the upstream DTS users and I thought all of them
> are fixed since long time (half a year) or even correct from the
> beginning. Now I found one more case with incorrect level, which I will fix.

That's just upstream, what about any downstream users?

> > I remain deeply unconvinced that remapping active low outputs like this
> > in the GPIO API is helping.

> The code is mapping them to correct state. The previous state was
> incorrect and did not allow to handle active high (which can happen).
> This is the effort to make code correct - driver and DTS.

We could handle inversions through an explicit property if that were
needed, that would be a less problematic transition and clearer in the
consumer code.

Download attachment "signature.asc" of type "application/pgp-signature" (489 bytes)