Date:   Wed, 8 Feb 2023 12:50:56 +0100
From:   Thierry Reding <thierry.reding@gmail.com>
To:     Manikanta Maddireddy <mmaddireddy@nvidia.com>
Cc:     bhelgaas@google.com, petlozup@nvidia.com,
        rafael.j.wysocki@intel.com, lpieralisi@kernel.org, robh@kernel.org,
        jeffy.chen@rock-chips.com, krzysztof.kozlowski+dt@linaro.org,
        jonathanh@nvidia.com, dmitry.osipenko@collabora.com,
        viresh.kumar@linaro.org, gregkh@linuxfoundation.org,
        steven.price@arm.com, kw@linux.com, linux-pci@vger.kernel.org,
        devicetree@vger.kernel.org, linux-kernel@vger.kernel.org,
        linux-tegra@vger.kernel.org, linux-pm@vger.kernel.org,
        vidyas@nvidia.com
Subject: Re: [RFC,v14 3/5] PCI / PM: Add support for the PCIe WAKE# signal
 for OF
Message-ID: <Y+OMoKOfduwj2PYp@orome>
References: <20230208111645.3863534-1-mmaddireddy@nvidia.com>
 <20230208111645.3863534-4-mmaddireddy@nvidia.com>
MIME-Version: 1.0
Content-Type: multipart/signed; micalg=pgp-sha256;
        protocol="application/pgp-signature"; boundary="1q5Qd0tqx26fwrBX"
Content-Disposition: inline
In-Reply-To: <20230208111645.3863534-4-mmaddireddy@nvidia.com>
User-Agent: Mutt/2.2.9 (2022-11-12)
Precedence: bulk


--1q5Qd0tqx26fwrBX
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
Content-Transfer-Encoding: quoted-printable

On Wed, Feb 08, 2023 at 04:46:43PM +0530, Manikanta Maddireddy wrote:
> From: Jeffy Chen <jeffy.chen@rock-chips.com>
>=20
> Add of_pci_setup_wake_irq() to parse the PCIe WAKE# interrupt from the
> device tree and set the wake irq. Add of_pci_teardown_wake_irq() to clear
> the wake irq.
>=20
> Call of_pci_setup_wake_irq() in pci_device_probe() to setup PCIe WAKE#
> interrupt during PCIe Endpoint enumeration.
>=20
> Enable or disable PCIe WAKE# interrupt in platform_pci_set_wakeup().
>=20
> Signed-off-by: Jeffy Chen <jeffy.chen@rock-chips.com>
> Signed-off-by: Manikanta Maddireddy <mmaddireddy@nvidia.com>
> ---
>=20
> Changes in v14:
> pci_platform_pm_ops structure is removed in latest kernel, so dropped
> pci-of driver. Instead, enable wake in platform_pci_set_wakeup().
>=20
> Changes in v13:
> Fix compiler error reported by kbuild test robot <fengguang.wu@intel.com>
>=20
> Changes in v12:
> Enable the wake irq in noirq stage to avoid possible irq storm.
>=20
> Changes in v11:
> Only support 1-per-device PCIe WAKE# pin as suggested.
>=20
> Changes in v10:
> Use device_set_wakeup_capable() instead of device_set_wakeup_enable(),
> since dedicated wakeirq will be lost in device_set_wakeup_enable(false).
>=20
> Changes in v9:
> Fix check error in .cleanup().
> Move dedicated wakeirq setup to setup() callback and use
> device_set_wakeup_enable() to enable/disable.
>=20
> Changes in v8:
> Add pci-of.c and use platform_pm_ops to handle the PCIe WAKE# signal.
>=20
> Changes in v7:
> Move PCIE_WAKE handling into pci core.
>=20
> Changes in v6:
> Fix device_init_wake error handling, and add some comments.
>=20
> Changes in v5:
> Rebase.
>=20
> Changes in v3:
> Fix error handling.
>=20
> Changes in v2:
> Use dev_pm_set_dedicated_wake_irq.
>=20
>  drivers/pci/of.c         | 49 ++++++++++++++++++++++++++++++++++++++++
>  drivers/pci/pci-driver.c | 10 ++++++++
>  drivers/pci/pci.c        |  7 ++++++
>  drivers/pci/pci.h        |  8 +++++++
>  4 files changed, 74 insertions(+)
>=20
> diff --git a/drivers/pci/of.c b/drivers/pci/of.c
> index ff897c40ed71..1c348e63f175 100644
> --- a/drivers/pci/of.c
> +++ b/drivers/pci/of.c
> @@ -13,6 +13,7 @@
>  #include <linux/of_irq.h>
>  #include <linux/of_address.h>
>  #include <linux/of_pci.h>
> +#include <linux/pm_wakeirq.h>
>  #include "pci.h"
> =20
>  #ifdef CONFIG_PCI
> @@ -705,3 +706,51 @@ u32 of_pci_get_slot_power_limit(struct device_node *=
node,
>  	return slot_power_limit_mw;
>  }
>  EXPORT_SYMBOL_GPL(of_pci_get_slot_power_limit);
> +
> +int of_pci_setup_wake_irq(struct pci_dev *pdev)
> +{
> +	struct pci_dev *ppdev;

Perhaps "parent" since that's what it is referring to? ppdev is a bit
vague.

> +	struct device_node *dn;
> +	int ret, irq;
> +
> +	/* Get the pci_dev of our parent. Hopefully it's a port. */
> +	ppdev =3D pdev->bus->self;
> +	/* Nope, it's a host bridge. */
> +	if (!ppdev)
> +		return 0;
> +
> +	dn =3D pci_device_to_OF_node(ppdev);
> +	if (!dn)
> +		return 0;
> +
> +	irq =3D of_irq_get_byname(dn, "wakeup");
> +	if (irq =3D=3D -EPROBE_DEFER) {
> +		return irq;
> +	} else if (irq < 0) {
> +		/* Ignore other errors, since a missing wakeup is non-fatal. */
> +		dev_info(&pdev->dev, "cannot get wakeup interrupt: %d\n", irq);

dev_dbg() maybe? As it is this would add an annoying info message for
basically every PCI controller on every DT-based board out there.

Thierry

--1q5Qd0tqx26fwrBX
Content-Type: application/pgp-signature; name="signature.asc"

-----BEGIN PGP SIGNATURE-----
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=ULUl
-----END PGP SIGNATURE-----

--1q5Qd0tqx26fwrBX--