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;
MIME-Version: 1.0
References: <20220831081603.3415-1-rrichter@amd.com> <20220831081603.3415-7-rrichter@amd.com>
In-Reply-To: <20220831081603.3415-7-rrichter@amd.com>
From:   "Rafael J. Wysocki" <rafael@kernel.org>
Date:   Thu, 8 Sep 2022 15:04:51 +0200
Message-ID: <CAJZ5v0gQqyFZHxe9fLX9WrMGBYhdna7Cpnx6bwTPFffy_koE5Q@mail.gmail.com>
Subject: Re: [PATCH 06/15] PCI/ACPI: Link host bridge to its ACPI fw node
To:     Robert Richter <rrichter@amd.com>
Cc:     Alison Schofield <alison.schofield@intel.com>,
        Vishal Verma <vishal.l.verma@intel.com>,
        Ira Weiny <ira.weiny@intel.com>,
        Ben Widawsky <bwidawsk@kernel.org>,
        Dan Williams <dan.j.williams@intel.com>,
        Bjorn Helgaas <bhelgaas@google.com>,
        "Rafael J. Wysocki" <rafael@kernel.org>, linux-cxl@vger.kernel.org,
        Linux Kernel Mailing List <linux-kernel@vger.kernel.org>,
        Len Brown <lenb@kernel.org>,
        Linux PCI <linux-pci@vger.kernel.org>,
        ACPI Devel Maling List <linux-acpi@vger.kernel.org>
Content-Type: text/plain; charset="UTF-8"
Precedence: bulk

On Wed, Aug 31, 2022 at 10:17 AM Robert Richter <rrichter@amd.com> wrote:
>
> A lookup of a host bridge's corresponding acpi device (struct
> acpi_device) is not possible, for example:
>
>         adev = ACPI_COMPANION(&host_bridge->dev);

Hmm.

x86 has this code in pcibios_root_bridge_prepare():

    if (!bridge->dev.parent) {
        struct pci_sysdata *sd = bridge->bus->sysdata;
        ACPI_COMPANION_SET(&bridge->dev, sd->companion);
    }

which should set the ACPI companion for the host bridge.

Moreover, on my x86 desktop /sys/devices/pci0000\:00/ (which is the
host bridge AFAICS) has

firmware_node -> ../LNXSYSTM:00/LNXSYBUS:00/PNP0A08:00

so clearly the ACPI companion is there.

Are we talking about ARM64 here?

> This could be useful to find a host bridge's fwnode handle and to
> determine and call additional host bridge ACPI parameters and methods
> such as HID/CID or _UID.
>
> Make this work by linking the host bridge to its ACPI fw node.
>
> Signed-off-by: Robert Richter <rrichter@amd.com>
> ---
>  drivers/acpi/pci_root.c | 1 +
>  1 file changed, 1 insertion(+)
>
> diff --git a/drivers/acpi/pci_root.c b/drivers/acpi/pci_root.c
> index d57cf8454b93..846c979e4c29 100644
> --- a/drivers/acpi/pci_root.c
> +++ b/drivers/acpi/pci_root.c
> @@ -1083,6 +1083,7 @@ struct pci_bus *acpi_pci_root_create(struct acpi_pci_root *root,
>                 goto out_release_info;
>
>         host_bridge = to_pci_host_bridge(bus->bridge);
> +       host_bridge->dev.fwnode = acpi_fwnode_handle(device);

So this would be replacing the existing mechanism on x86, right?

>         if (!(root->osc_control_set & OSC_PCI_EXPRESS_NATIVE_HP_CONTROL))
>                 host_bridge->native_pcie_hotplug = 0;
>         if (!(root->osc_control_set & OSC_PCI_SHPC_NATIVE_HP_CONTROL))
> --