Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 2620:137:e000::1:20 as permitted sender) client-ip=2620:137:e000::1:20;
Date:   Mon, 31 Oct 2022 11:50:38 -0700
From:   Zev Weiss <zev@bewilderbeest.net>
To:     Mark Brown <broonie@kernel.org>
Cc:     Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org>,
        Rob Herring <robh@kernel.org>,
        Liam Girdwood <lgirdwood@gmail.com>,
        Krzysztof Kozlowski <krzysztof.kozlowski+dt@linaro.org>,
        devicetree@vger.kernel.org, linux-kernel@vger.kernel.org,
        Naresh Solanki <naresh.solanki@9elements.com>,
        Patrick Rudolph <patrick.rudolph@9elements.com>,
        Laxman Dewangan <ldewangan@nvidia.com>,
        openbmc@lists.ozlabs.org
Subject: Re: [PATCH 2/3] dt-bindings: regulator: Add regulator-output bindingg
Message-ID: <Y2AY/tQLQ6m+SXI7@hatter.bewilderbeest.net>
References: <20220925220319.12572-1-zev@bewilderbeest.net>
 <20220925220319.12572-3-zev@bewilderbeest.net>
 <20220929210714.GA2684335-robh@kernel.org>
 <YzYNt+IQRomycRLs@hatter.bewilderbeest.net>
 <Y1rRCq9Kdd2zPPkw@hatter.bewilderbeest.net>
 <ee37b5a1-5afc-71b3-f777-add295d9ce17@linaro.org>
 <Y1tWpikPogEtV0+x@hatter.bewilderbeest.net>
 <Y1v6migO2PNV4ksW@sirena.org.uk>
 <Y1wxMk2x25AeRwLr@hatter.bewilderbeest.net>
 <Y1/tnic0qc/Ll/5u@sirena.org.uk>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii; format=flowed
Content-Disposition: inline
In-Reply-To: <Y1/tnic0qc/Ll/5u@sirena.org.uk>
Precedence: bulk

On Mon, Oct 31, 2022 at 08:45:34AM PDT, Mark Brown wrote:
>On Fri, Oct 28, 2022 at 12:44:50PM -0700, Zev Weiss wrote:
>> On Fri, Oct 28, 2022 at 08:51:54AM PDT, Mark Brown wrote:
>
>> > We don't turn things off on reboot?  We don't do anything in particular
>> > on reboot...
>
>> Okay, perhaps not on reboot specifically, but the userspace-consumer driver
>> has a regulator_bulk_disable() in its .remove function, so it would be
>> triggered at least by a module unload (which is sort of why I ended up with
>> the "when software relinquishes control" wording in the patch).  If we're
>> going to continue with the plan of using that driver for this functionality
>> (which seems overall quite reasonable to me), we need a way to express that
>> that must not happen on this hardware.
>
>Ah, that would be the test driver not intended to be used in production
>then...  That shouldn't be a blocker for the DT binding, and if there's
>a different compatible string for this application then we can either
>make the userspace consumer do something different based on that
>compatible string or have a new driver which does something more
>sensible and perhaps has a better userspace ABI.  Either way so long as
>we can tell the thing being described is a BMC output from the DT
>binding I think we can leave it up to the OS to do something constructive
>with that rather than trying to control the specific behaviour in the
>binding.

Ah, alright -- that seems like a nice (obvious in retrospect, of course) 
solution that should work well I think.  I'll post a v2 with that 
approach soon.


Thanks,
Zev