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:   Tue, 7 Jun 2022 09:07:59 +0200
From:   =?UTF-8?B?Q2zDqW1lbnQgTMOpZ2Vy?= <clement.leger@bootlin.com>
To:     Alan Stern <stern@rowland.harvard.edu>
Cc:     Greg Kroah-Hartman <gregkh@linuxfoundation.org>,
        Nicolas Ferre <nicolas.ferre@microchip.com>,
        Alexandre Belloni <alexandre.belloni@bootlin.com>,
        Claudiu Beznea <claudiu.beznea@microchip.com>,
        linux-usb@vger.kernel.org, linux-arm-kernel@lists.infradead.org,
        linux-kernel@vger.kernel.org
Subject: Re: [PATCH] usb: host: ohci-at91: add support to enter suspend
 using SMC
Message-ID: <20220607090759.3fc0b003@fixe.home>
In-Reply-To: <Yp5DpPpW5/3SnuJl@rowland.harvard.edu>
References: <20220606141802.165252-1-clement.leger@bootlin.com>
        <Yp5DpPpW5/3SnuJl@rowland.harvard.edu>
Organization: Bootlin
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: quoted-printable
Precedence: bulk

Le Mon, 6 Jun 2022 14:12:52 -0400,
Alan Stern <stern@rowland.harvard.edu> a =C3=A9crit :

> On Mon, Jun 06, 2022 at 04:18:02PM +0200, Cl=C3=A9ment L=C3=A9ger wrote:
> > When Linux is running under OP-TEE, the SFR is set as secured and thus
> > the AT91_OHCIICR_USB_SUSPEND register isn't accessible. Add a SMC to
> > do the appropriate call to suspend the controller.
> > The SMC id is fetched from the device-tree property
> > "microchip,suspend-smc-id". if present, then the syscon regmap is not
> > used to enter suspend and a SMC is issued.
> >=20
> > Signed-off-by: Cl=C3=A9ment L=C3=A9ger <clement.leger@bootlin.com>
> > --- =20
>=20
> Acked-by: Alan Stern <stern@rowland.harvard.edu>
>=20
> However, this is a little weird...  You've written=20
> usb_hcd_at91_probe() so that the SMC is detected in preference to the=20
> regmap, but then you wrote ohci_at91_port_suspend() so that the regmap=20
> is used in preference to the SMC.  It's not wrong, but it is confusing=20
> to read.
>=20
> Do you want to rewrite the patch to make the two routines agree on which=
=20
> mechanism to use by default?
>=20
> Alan Stern

Hi Alan,

I'll rewrite that ! I did it in this specific order in the probe to
allow overloading the device-tree with a SMC ID without removing the
syscon property. This way, the regmap stays the default if no
"microchip,suspend-smc-id" property is provided.

Does it sounds good to you ?

Thanks,

--=20
Cl=C3=A9ment L=C3=A9ger,
Embedded Linux and Kernel engineer at Bootlin
https://bootlin.com