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:   Sun, 30 Apr 2023 18:43:11 +0100
From:   Jonathan Cameron <jic23@kernel.org>
To:     Nuno =?UTF-8?B?U8Oh?= <noname.nuno@gmail.com>
Cc:     Masahiro Honda <honda@mechatrax.com>,
        Lars-Peter Clausen <lars@metafoo.de>,
        Michael Hennerich <Michael.Hennerich@analog.com>,
        linux-iio@vger.kernel.org, linux-kernel@vger.kernel.org
Subject: Re: [PATCH v3] Fix IRQ issue by setting IRQ_DISABLE_UNLAZY flag
Message-ID: <20230430184311.7443e1d7@jic23-huawei>
In-Reply-To: <b6a96cca17832219042c90baec2442b1a5bad7b7.camel@gmail.com>
References: <20230420102316.757-1-honda@mechatrax.com>
        <20230423121528.259d3cd4@jic23-huawei>
        <b6a96cca17832219042c90baec2442b1a5bad7b7.camel@gmail.com>
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: quoted-printable
Precedence: bulk

On Mon, 24 Apr 2023 11:09:34 +0200
Nuno S=C3=A1 <noname.nuno@gmail.com> wrote:

> Hi Jonathan,
>=20
> On Sun, 2023-04-23 at 12:15 +0100, Jonathan Cameron wrote:
> > On Thu, 20 Apr 2023 19:23:16 +0900
> > Masahiro Honda <honda@mechatrax.com> wrote:
> >  =20
> > > The Sigma-Delta ADCs supported by this driver can use SDO as an inter=
rupt
> > > line to indicate the completion of a conversion. However, some devices
> > > cannot properly detect the completion of a conversion by an interrupt.
> > > This is for the reason mentioned in the following commit.
> > >=20
> > > commit e9849777d0e2 ("genirq: Add flag to force mask in
> > > =C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=
=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0 disable_irq[_nosy=
nc]()")
> > >=20
> > > A read operation is performed by an extra interrupt before the comple=
tion
> > > of a conversion. This patch fixes the issue by setting IRQ_DISABLE_UN=
LAZY
> > > flag.
> > >=20
> > > Signed-off-by: Masahiro Honda <honda@mechatrax.com>
> > > ---
> > > v3:
> > > =C2=A0- Remove the Kconfig option.
> > > v2:
> > > https://lore.kernel.org/linux-iio/20230414102744.150-1-honda@mechatra=
x.com/
> > > =C2=A0- Rework commit message.
> > > =C2=A0- Add a new entry in the Kconfig.
> > > =C2=A0- Call irq_clear_status_flags(irq, IRQ_DISABLE_UNLAZY) when fre=
eing the
> > > IRQ.
> > > v1:
> > > https://lore.kernel.org/linux-iio/20230306044737.862-1-honda@mechatra=
x.com/
> > >=20
> > > =C2=A0drivers/iio/adc/ad_sigma_delta.c | 25 ++++++++++++++++++++-----
> > > =C2=A01 file changed, 20 insertions(+), 5 deletions(-)
> > >=20
> > > diff --git a/drivers/iio/adc/ad_sigma_delta.c
> > > b/drivers/iio/adc/ad_sigma_delta.c
> > > index d8570f620..215ecbedb 100644
> > > --- a/drivers/iio/adc/ad_sigma_delta.c
> > > +++ b/drivers/iio/adc/ad_sigma_delta.c
> > > @@ -565,6 +565,14 @@ int ad_sd_validate_trigger(struct iio_dev *indio=
_dev,
> > > struct iio_trigger *trig)
> > > =C2=A0}
> > > =C2=A0EXPORT_SYMBOL_NS_GPL(ad_sd_validate_trigger, IIO_AD_SIGMA_DELTA=
);
> > > =C2=A0
> > > +static void ad_sd_free_irq(void *sd)
> > > +{
> > > +=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0struct ad_sigma_delta *sig=
ma_delta =3D sd;
> > > +
> > > +=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0irq_clear_status_flags(sig=
ma_delta->spi->irq, IRQ_DISABLE_UNLAZY);
> > > +=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0free_irq(sigma_delta->spi-=
>irq, sigma_delta);
> > > +} =20
> >=20
> > Don't fuse the two operations unwinding like this.=C2=A0 Just register =
a callback
> > that only
> > does the irq_clear_status_flags immediately after setting them.=C2=A0 T=
hen leave =20
>=20
> I was the one to propose fusing them together because I thought that we c=
ould
> have issues by clearing the flag after calling free_irq(). After looking =
again
> at the IRQ code, I can see that it is not up to free_irq() to free the al=
located
> irq_descs (that might only happen when unmapping the virq) which means we=
 should
> be fine doing the normal way.

Ah. I'd missed the ordering.  If that had been valid (and I think you are c=
orrect
that it is not required) then a comment to make that clear would be necessa=
ry.

Usual case of: When doing something non obvious with ordering, say why.

>=20
> That said, looking at the only users that care to clear this flag, it loo=
ks like
> they do it before calling free_irq(). Hence, I'm not sure if there's anyt=
hing
> subtle going on. In fact, looking at this line:
>=20
> https://elixir.bootlin.com/linux/latest/source/kernel/irq/manage.c#L1909
>=20
> I'm not so sure we actually need to clear the flag as for these devices, =
we
> should only have one consumer/action per IRQ. Anyways, probably for corre=
ctness
> we should still explicitly clear it?

Good question...  Looks to me like a driver shouldn't be clearing this flag
itself, but it's probably harmless in most cases.

I'd drop the clear of the status flag, perhaps adding a comment that
the irq core does it for us.

Jonathan


>=20
> - Nuno S=C3=A1
>=20