[U-Boot] [PATCH v5 2/4] net: designware: Add support to PCI designware devices

[Simon Glass]

On 15 September 2015 at 08:20, Bin Meng <bmeng.cn at gmail.com> wrote:
> Hi Simon,
>
> On Tue, Sep 15, 2015 at 9:51 PM, Simon Glass <sjg at chromium.org> wrote:
>> On 11 September 2015 at 04:24, Bin Meng <bmeng.cn at gmail.com> wrote:
>>> The Designware ethernet controller is also seen on PCI bus, e.g.
>>> on Intel Quark SoC. Add this support in the DM version driver.
>>>
>>> Signed-off-by: Bin Meng <bmeng.cn at gmail.com>
>>>
>>> ---
>>>
>>> Changes in v5:
>>> - Wrap PCI device support with CONFIG_DM_PCI
>>>
>>> Changes in v3:
>>> - Change to use dm_pci_read_config32()
>>>
>>> Changes in v2:
>>> - Change to use device_is_on_pci_bus()
>>>
>>>  drivers/net/designware.c | 42 ++++++++++++++++++++++++++++++++++++++++++
>>>  1 file changed, 42 insertions(+)
>>
>> Acked-by: Simon Glass <sjg at chromium.org>
>>
>> Please see below.
>>
>>>
>>> diff --git a/drivers/net/designware.c b/drivers/net/designware.c
>>> index ae78d21..6433896 100644
>>> --- a/drivers/net/designware.c
>>> +++ b/drivers/net/designware.c
>>> @@ -14,6 +14,7 @@
>>>  #include <errno.h>
>>>  #include <miiphy.h>
>>>  #include <malloc.h>
>>> +#include <pci.h>
>>>  #include <linux/compiler.h>
>>>  #include <linux/err.h>
>>>  #include <asm/io.h>
>>> @@ -558,6 +559,22 @@ static int designware_eth_write_hwaddr(struct udevice *dev)
>>>         return _dw_write_hwaddr(priv, pdata->enetaddr);
>>>  }
>>>
>>> +static int designware_eth_bind(struct udevice *dev)
>>> +{
>>> +#ifdef CONFIG_DM_PCI
>>> +       static int num_cards;
>>> +       char name[20];
>>> +
>>> +       /* Create a unique device name for PCI type devices */
>>> +       if (device_is_on_pci_bus(dev)) {
>>> +               sprintf(name, "eth_designware#%u", num_cards++);
>>> +               device_set_name(dev, name);
>>> +       }
>>> +#endif
>>> +
>>> +       return 0;
>>> +}
>>> +
>>>  static int designware_eth_probe(struct udevice *dev)
>>>  {
>>>         struct eth_pdata *pdata = dev_get_platdata(dev);
>>> @@ -565,6 +582,23 @@ static int designware_eth_probe(struct udevice *dev)
>>>         u32 iobase = pdata->iobase;
>>>         int ret;
>>>
>>> +#ifdef CONFIG_DM_PCI
>>> +       /*
>>> +        * If we are on PCI bus, either directly attached to a PCI root port,
>>> +        * or via a PCI bridge, fill in platdata before we probe the hardware.
>>> +        */
>>> +       if (device_is_on_pci_bus(dev)) {
>>> +               pci_dev_t bdf = pci_get_bdf(dev);
>>> +
>>> +               dm_pci_read_config32(dev, PCI_BASE_ADDRESS_0, &iobase);
>>> +               iobase &= PCI_BASE_ADDRESS_MEM_MASK;
>>> +               iobase = pci_mem_to_phys(bdf, iobase);
>>> +
>>> +               pdata->iobase = iobase;
>>> +               pdata->phy_interface = PHY_INTERFACE_MODE_RMII;
>>> +       }
>>> +#endif
>>> +
>>>         debug("%s, iobase=%x, priv=%p\n", __func__, iobase, priv);
>>>         priv->mac_regs_p = (struct eth_mac_regs *)iobase;
>>>         priv->dma_regs_p = (struct eth_dma_regs *)(iobase + DW_DMA_BASE_OFFSET);
>>> @@ -617,10 +651,18 @@ U_BOOT_DRIVER(eth_designware) = {
>>>         .id     = UCLASS_ETH,
>>>         .of_match = designware_eth_ids,
>>>         .ofdata_to_platdata = designware_eth_ofdata_to_platdata,
>>> +       .bind   = designware_eth_bind,
>>>         .probe  = designware_eth_probe,
>>>         .ops    = &designware_eth_ops,
>>>         .priv_auto_alloc_size = sizeof(struct dw_eth_dev),
>>>         .platdata_auto_alloc_size = sizeof(struct eth_pdata),
>>>         .flags = DM_FLAG_ALLOC_PRIV_DMA,
>>>  };
>>> +
>>> +static struct pci_device_id supported[] = {
>>> +       { PCI_DEVICE(PCI_VENDOR_ID_INTEL, PCI_DEVICE_ID_INTEL_QRK_EMAC) },
>>> +       { }
>>
>> Rather than ending up with a table of these device IDs, should this go
>> in the device tree?
>
> I am OK for either way. In fact compared to the widely available PCI
> NS16550 devices from many chipset vendors, there are just two or three
> PCI variants of designware ethernet devices so far (seen from linux
> driver). I guess putting a device ID table here is not that bad.

OK let's revisit it if needed.

Applied to u-boot-x86, thanks!