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:   Wed, 31 Aug 2022 12:58:30 +0100
From:   Jonathan Cameron <Jonathan.Cameron@huawei.com>
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>,
        <linux-cxl@vger.kernel.org>, <linux-kernel@vger.kernel.org>,
        Bjorn Helgaas <bhelgaas@google.com>,
        "Rafael J. Wysocki" <rafael@kernel.org>,
        Len Brown <lenb@kernel.org>
Subject: Re: [PATCH 12/15] cxl/acpi: Skip devm_cxl_port_enumerate_dports()
 when in RCD mode
Message-ID: <20220831125830.00005105@huawei.com>
In-Reply-To: <20220831081603.3415-13-rrichter@amd.com>
References: <20220831081603.3415-1-rrichter@amd.com>
        <20220831081603.3415-13-rrichter@amd.com>
MIME-Version: 1.0
Content-Type: text/plain; charset="US-ASCII"
Content-Transfer-Encoding: 7bit
Precedence: bulk

On Wed, 31 Aug 2022 10:16:00 +0200
Robert Richter <rrichter@amd.com> wrote:

> RCD mode has a different enumeration scheme other than in CXL VH mode.
> An RCD is directly connected to an RCH without downstream and upstream
> ports showing up in between in the PCI hierarchy. Skip dport
> enumeration for RCHs. Upstream and downstream ports of RCH and RCD
> will be setup separately in a later patch.
> 
> Introduce the function is_rch_uport() to detect an RCH port. For RCHs
> the parent root port is not the "ACPI0017" device and instead does not
> have a fw node connected to it.
> 
> Signed-off-by: Robert Richter <rrichter@amd.com>
> ---
>  drivers/cxl/core/pci.c | 20 +++++++++++++++++++-
>  1 file changed, 19 insertions(+), 1 deletion(-)
> 
> diff --git a/drivers/cxl/core/pci.c b/drivers/cxl/core/pci.c
> index 0dbbe8d39b07..86ed112eb262 100644
> --- a/drivers/cxl/core/pci.c
> +++ b/drivers/cxl/core/pci.c
> @@ -65,6 +65,15 @@ static int match_add_dports(struct pci_dev *pdev, void *data)
>  	return 0;
>  }
>  
> +/*
> + * A parent of an RCH (CXL 1.1 host) is a plain platform device while
> + * a 2.0 host links to the ACPI0017 root device.
> + */
> +static inline bool is_rch_uport(struct cxl_port *port)
> +{
> +	return is_cxl_port(&port->dev) && !port->dev.parent->fwnode;

I'm not keen on the presence of fwnode being used to distinguish anything.
That's the sort of thing that gets 'fixed' by later patches.

Can we check something more specific?

> +}
> +
>  /**
>   * devm_cxl_port_enumerate_dports - enumerate downstream ports of the upstream port
>   * @port: cxl_port whose ->uport is the upstream of dports to be enumerated
> @@ -74,10 +83,19 @@ static int match_add_dports(struct pci_dev *pdev, void *data)
>   */
>  int devm_cxl_port_enumerate_dports(struct cxl_port *port)
>  {
> -	struct pci_bus *bus = cxl_port_to_pci_bus(port);
> +	struct pci_bus *bus;
>  	struct cxl_walk_context ctx;
>  	int type;
>  
> +	/*
> +	 * Skip enumeration in Restricted CXL Device mode as the
> +	 * device has been already registered at the host's dport
> +	 * during host discovery.
> +	 */
> +	if (is_rch_uport(port))
> +		return 0;
> +
> +	bus = cxl_port_to_pci_bus(port);
>  	if (!bus)
>  		return -ENXIO;
>