Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 2620:137:e000::3:1 as permitted sender) client-ip=2620:137:e000::3:1;
MIME-Version: 1.0
References: <20230918080314.11959-1-jagathjog1996@gmail.com>
 <20230918080314.11959-2-jagathjog1996@gmail.com> <20230924143710.7c6edc4a@jic23-huawei>
 <CAM+2EuJ8o5X2ucph8gic2=03bbLQmUCX=j3SFLrqD4Y6rwXs9A@mail.gmail.com>
 <20230930170530.0b8f185c@jic23-huawei> <CAM+2EuKzBVbTF2SrpYEYfzap1wrONboFV-QuTU9Fz7sVjqJLeA@mail.gmail.com>
 <20231010100002.0163d681@jic23-huawei> <CACRpkdY+K90mN1Q1tf38FLRgEsz3q8dK9SJYSQVwGe=PL3FaUQ@mail.gmail.com>
 <20231010154216.6032a1c2@jic23-huawei>
In-Reply-To: <20231010154216.6032a1c2@jic23-huawei>
From:   Linus Walleij <linus.walleij@linaro.org>
Date:   Tue, 10 Oct 2023 21:51:17 +0200
Message-ID: <CACRpkdY4yJKmc_Sbr=4sJ0WseCEEBOQ7wKj4fBUMG2_cx-Xxbw@mail.gmail.com>
Subject: Re: [RFC 1/2] dt-bindings: iio: imu: Add DT binding doc for BMI323
To:     Jonathan Cameron <jic23@kernel.org>
Cc:     Jagath Jog J <jagathjog1996@gmail.com>,
        andriy.shevchenko@linux.intel.com, lars@metafoo.de,
        robh+dt@kernel.org, krzysztof.kozlowski+dt@linaro.org,
        linux-iio@vger.kernel.org, devicetree@vger.kernel.org,
        linux-kernel@vger.kernel.org
Content-Type: text/plain; charset="UTF-8"
Content-Transfer-Encoding: quoted-printable
Precedence: bulk

On Tue, Oct 10, 2023 at 4:42=E2=80=AFPM Jonathan Cameron <jic23@kernel.org>=
 wrote:

> We kind of lost the question along the way.  Wasn't so much about whether
> there was a generic binding but more about whether it is worth providing
> separate controls for the two IRQ pins?  Or just assume no one is crazy
> enough to play that level of mix and match.

Ugh no, that's upfront design for a nonexistent use case.

- First, to even consider open drain the designer need to be really
  short of IRQ lines/rails, and, despite knowing it's a bad idea, decide
  to share this line between several peripherals, even though it will
  require I2C traffic to just determine which one even fired the IRQ.

- Second, be interested in using two IRQs to distinguish between
  different events? When we just faced the situation that we had
  too few IRQ lines so we need to start sharing them with open
  drain...?

It's not gonna happen.

Stay with just drive-open-drain; and configure them all as that if
that property is set.

Yours,
Linus Walleij