Re: [PATCH 04/11] usb: xhci: Add Intel cherrytrail extended cap / otg phy mux handling

[Heikki Krogerus <heikki.krogerus@linux.intel.com>]

Hi,

On Fri, Sep 01, 2017 at 11:48:38PM +0200, Hans de Goede wrote:
> The Intel cherrytrail xhci controller has an extended cap mmio-range
> which contains registers to control the muxing to the xhci (host mode)
> or the dwc3 (device mode) and vbus-detection for the otg usb-phy.
> 
> Having a mux driver included in the xhci code (or under drivers/usb/host)
> is not desirable. So this commit adds a simple handler for this extended
> capability, which creates a platform device with the caps mmio region as
> resource, this allows us to write a separate platform mux driver for the
> mux.
> 
> Signed-off-by: Hans de Goede <hdegoede@redhat.com>
> ---
> Changes in v2:
> -Check xHCI controller PCI device-id instead of only checking for the
>  Intel Extended capability ID, as the Extended capability ID is used on
>  other model Intel xHCI controllers too
> ---
>  drivers/usb/host/Makefile            |  2 +-
>  drivers/usb/host/xhci-intel-quirks.c | 85 ++++++++++++++++++++++++++++++++++++
>  drivers/usb/host/xhci-pci.c          |  4 ++
>  drivers/usb/host/xhci.h              |  2 +
>  4 files changed, 92 insertions(+), 1 deletion(-)
>  create mode 100644 drivers/usb/host/xhci-intel-quirks.c
> 
> diff --git a/drivers/usb/host/Makefile b/drivers/usb/host/Makefile
> index cf2691fffcc0..441edf82eb1c 100644
> --- a/drivers/usb/host/Makefile
> +++ b/drivers/usb/host/Makefile
> @@ -63,7 +63,7 @@ obj-$(CONFIG_USB_OHCI_HCD_DAVINCI)	+= ohci-da8xx.o
>  obj-$(CONFIG_USB_UHCI_HCD)	+= uhci-hcd.o
>  obj-$(CONFIG_USB_FHCI_HCD)	+= fhci.o
>  obj-$(CONFIG_USB_XHCI_HCD)	+= xhci-hcd.o
> -obj-$(CONFIG_USB_XHCI_PCI)	+= xhci-pci.o
> +obj-$(CONFIG_USB_XHCI_PCI)	+= xhci-pci.o xhci-intel-quirks.o
>  obj-$(CONFIG_USB_XHCI_PLATFORM) += xhci-plat-hcd.o
>  obj-$(CONFIG_USB_XHCI_MTK)	+= xhci-mtk.o
>  obj-$(CONFIG_USB_XHCI_TEGRA)	+= xhci-tegra.o
> diff --git a/drivers/usb/host/xhci-intel-quirks.c b/drivers/usb/host/xhci-intel-quirks.c
> new file mode 100644
> index 000000000000..819f5f9da9ee
> --- /dev/null
> +++ b/drivers/usb/host/xhci-intel-quirks.c
> @@ -0,0 +1,85 @@
> +/*
> + * Intel Vendor Defined XHCI extended capability handling
> + *
> + * Copyright (c) 2016) Hans de Goede <hdegoede@redhat.com>
> + *
> + * Loosely based on android x86 kernel code which is:
> + *
> + * Copyright (C) 2014 Intel Corp.
> + *
> + * Author: Wu, Hao
> + *
> + * This program is free software; you can redistribute it and/or modify
> + * it under the terms of the GNU General Public License version 2 as
> + * published by the Free Software Foundation.
> + *
> + * This program is distributed in the hope that it will be useful, but
> + * WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY
> + * or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License
> + * for more details.
> + *
> + * You should have received a copy of the GNU General Public License
> + * along with this program;
> + */
> +
> +#include <linux/platform_device.h>
> +#include "xhci.h"
> +
> +/* Extended capability IDs for Intel Vendor Defined */
> +#define XHCI_EXT_CAPS_INTEL_HOST_CAP	192
> +
> +static void xhci_intel_unregister_pdev(void *arg)
> +{
> +	platform_device_unregister(arg);
> +}
> +
> +int xhci_create_intel_cht_mux_pdev(struct xhci_hcd *xhci)
> +{
> +	struct usb_hcd *hcd = xhci_to_hcd(xhci);
> +	struct device *dev = hcd->self.controller;
> +	struct platform_device *pdev;
> +	struct resource	res = { 0, };
> +	int ret, ext_offset;
> +
> +	ext_offset = xhci_find_next_ext_cap(&xhci->cap_regs->hc_capbase, 0,
> +					    XHCI_EXT_CAPS_INTEL_HOST_CAP);
> +	if (!ext_offset) {
> +		xhci_err(xhci, "couldn't find Intel ext caps\n");
> +		return -ENODEV;
> +	}
> +
> +	pdev = platform_device_alloc("intel_cht_usb_mux", PLATFORM_DEVID_NONE);
> +	if (!pdev) {
> +		xhci_err(xhci, "couldn't allocate intel_cht_usb_mux pdev\n");
> +		return -ENOMEM;
> +	}
> +
> +	res.start = hcd->rsrc_start + ext_offset;
> +	res.end	  = res.start + 0x3ff;
> +	res.name  = "intel_cht_usb_mux";
> +	res.flags = IORESOURCE_MEM;
> +
> +	ret = platform_device_add_resources(pdev, &res, 1);
> +	if (ret) {
> +		dev_err(dev, "couldn't add resources to intel_cht_usb_mux pdev\n");
> +		platform_device_put(pdev);
> +		return ret;
> +	}

Previously the problem with this was that since platform bus reserves
the memory region, usb core fails to register the hcd, as it also
wants to reserve the same memory region. So xhci with this failed to
probe.

So has that been fixed? Does xhci really get registered with this?

> +	pdev->dev.parent = dev;
> +
> +	ret = platform_device_add(pdev);
> +	if (ret) {
> +		dev_err(dev, "couldn't register intel_cht_usb_mux pdev\n");
> +		platform_device_put(pdev);
> +		return ret;
> +	}
> +
> +	ret = devm_add_action_or_reset(dev, xhci_intel_unregister_pdev, pdev);
> +	if (ret) {
> +		dev_err(dev, "couldn't add unregister action for intel_cht_usb_mux pdev\n");
> +		return ret;
> +	}
> +
> +	return 0;
> +}
> diff --git a/drivers/usb/host/xhci-pci.c b/drivers/usb/host/xhci-pci.c
> index 8071c8fdd15e..b55c1e96abf0 100644
> --- a/drivers/usb/host/xhci-pci.c
> +++ b/drivers/usb/host/xhci-pci.c
> @@ -188,6 +188,7 @@ static void xhci_pci_quirks(struct device *dev, struct xhci_hcd *xhci)
>  	if (pdev->vendor == PCI_VENDOR_ID_INTEL &&
>  		 pdev->device == PCI_DEVICE_ID_INTEL_CHERRYVIEW_XHCI) {
>  		xhci->quirks |= XHCI_SSIC_PORT_UNUSED;
> +		xhci->quirks |= XHCI_INTEL_CHT_USB_MUX;
>  	}
>  	if (pdev->vendor == PCI_VENDOR_ID_INTEL &&
>  	    (pdev->device == PCI_DEVICE_ID_INTEL_CHERRYVIEW_XHCI ||
> @@ -328,6 +329,9 @@ static int xhci_pci_probe(struct pci_dev *dev, const struct pci_device_id *id)
>  	if (xhci->quirks & XHCI_PME_STUCK_QUIRK)
>  		xhci_pme_acpi_rtd3_enable(dev);
>  
> +	if (xhci->quirks & XHCI_INTEL_CHT_USB_MUX)
> +		xhci_create_intel_cht_mux_pdev(xhci);
> +
>  	/* USB-2 and USB-3 roothubs initialized, allow runtime pm suspend */
>  	pm_runtime_put_noidle(&dev->dev);
>  
> diff --git a/drivers/usb/host/xhci.h b/drivers/usb/host/xhci.h
> index e3e935291ed6..f722ee31e50d 100644
> --- a/drivers/usb/host/xhci.h
> +++ b/drivers/usb/host/xhci.h
> @@ -1821,6 +1821,7 @@ struct xhci_hcd {
>  #define XHCI_LIMIT_ENDPOINT_INTERVAL_7	(1 << 26)
>  #define XHCI_U2_DISABLE_WAKE	(1 << 27)
>  #define XHCI_ASMEDIA_MODIFY_FLOWCONTROL	(1 << 28)
> +#define XHCI_INTEL_CHT_USB_MUX	(1 << 29)

Is that really necessary? Why not just register the platform device
the moment we find match for the PCI device ID?


Br,

-- 
heikki