On 13 December 2013 21:55, Marek Vasut <ma...@denx.de> wrote:
> The detection function of the EHCI PCI controller was really cryptic,
> add a beefy comment and clean the portion of the code up a bit. No
> change in the logic of the code.
>
> Signed-off-by: Marek Vasut <ma...@denx.de>
> Cc: Simon Glass <s...@chromium.org>
Acked-by: Simon Glass <s...@chromium.org>
> ---
> drivers/usb/host/ehci-pci.c | 28 +++++++++++++++++++++++++---
> 1 file changed, 25 insertions(+), 3 deletions(-)
>
> diff --git a/drivers/usb/host/ehci-pci.c b/drivers/usb/host/ehci-pci.c
> index 7a1ffe5..991b199 100644
> --- a/drivers/usb/host/ehci-pci.c
> +++ b/drivers/usb/host/ehci-pci.c
> @@ -54,9 +54,31 @@ static pci_dev_t ehci_find_class(int index)
> bdf += PCI_BDF(0, 0, 1)) {
> pci_read_config_dword(bdf, PCI_CLASS_REVISION,
> &class);
> - if ((class >> 8 == PCI_CLASS_SERIAL_USB_EHCI)
> - && !index--)
> - return bdf;
> + class >>= 8;
> + /*
> + * Here be dragons! In case we have multiple
> + * PCI EHCI controllers, this function will
> + * be called multiple times as well. This
> + * function will scan the PCI busses, always
> + * starting from bus 0, device 0, function 0,
> + * until it finds an USB controller. The USB
> + * stack gives us an 'index' of a controller
> + * that is currently being registered, which
> + * is a number, starting from 0 and growing
> + * in ascending order as controllers are
> added.
> + * To avoid probing the same controller in tne
> + * subsequent runs of this function, we will
> + * skip 'index - 1' detected controllers and
> + * report the index'th controller.
> + */
> + if (class != PCI_CLASS_SERIAL_USB_EHCI)
> + continue;
> + if (index) {
> + index--;
> + continue;
> + }
> + /* Return index'th controller. */
> + return bdf;
> }
> }
> }
> --
> 1.8.4.3
>
_______________________________________________
U-Boot mailing list
U-Boot@lists.denx.de
http://lists.denx.de/mailman/listinfo/u-boot
