On Thu, Sep 8, 2016 at 3:52 PM, Arnd Bergmann [off-list ref] wrote:

On Thursday, September 8, 2016 3:41:43 PM CEST Linus Walleij wrote:

quoted

The SMSC9112 ethernet controller is connected to chip select 2
on the EBI2 bus on the APQ8060 Dragonboard. We set this up by
activating EBI2, creating a chipselect entry as a subnode, and then
putting the ethernet controller in a subnode of the chipselect.

After the chipselect is configured, the SMSC device will be
instantiated.

Signed-off-by: Linus Walleij <redacted>
---

One more thing I noticed recently is that we still have some
hardcoded configuration in the driver:

  #ifdef CONFIG_ARCH_PXA
    #define SMC_USE_PXA_DMA       1
    #define SMC_USE_16BIT         0
    #define SMC_USE_32BIT         1
    #define SMC_IRQ_SENSE         IRQF_TRIGGER_FALLING
  #elif defined(CONFIG_SH_MAGIC_PANEL_R2)
    #define SMC_USE_16BIT         0
    #define SMC_USE_32BIT         1
    #define SMC_IRQ_SENSE         IRQF_TRIGGER_LOW
  #elif defined(CONFIG_ARCH_OMAP3)
    #define SMC_USE_16BIT         0
    #define SMC_USE_32BIT         1
    #define SMC_IRQ_SENSE         IRQF_TRIGGER_LOW
    #define SMC_MEM_RESERVED      1
  #elif defined(CONFIG_ARCH_OMAP2)
    #define SMC_USE_16BIT         0
    #define SMC_USE_32BIT         1
    #define SMC_IRQ_SENSE         IRQF_TRIGGER_LOW
    #define SMC_MEM_RESERVED      1
  #else
  /*
   * Default configuration
   */

  #define SMC_DYNAMIC_BUS_CONFIG
  #endif

and I suspect we have to convert the OMAP2/OMAP3 portion here to use
dynamic configuration, otherwise things may go wrong in a multiplatform
kernel.

Yes, but that is not the driver enabled by this patch. This is
compatible = "smsc,lan9221", "smsc,lan9115";
which has the driver smsc911x.c, but the problem
you're pointing to is in the driver smc911x.c.
or rather its header file.

One letter difference I know, the naming of these
drivers is not human-friendly. I make the mistake
all the time.

Yours,
Linus Walleij