On 09/21/10 18:26, Gadiyar, Anand wrote:
> On Tue, Sep 21, 2010 at 9:33 PM, Igor Grinberg <grinberg@xxxxxxxxxxxxxx> wrote:
>> add support for hsusb host ports 1, 2 and on-module usb hub.
>>
>> Signed-off-by: Igor Grinberg <grinberg@xxxxxxxxxxxxxx>
>> ---
>> arch/arm/mach-omap2/board-cm-t3517.c | 50 ++++++++++++++++++++++++++++++++++
> ...
>
>> @@ -100,6 +102,47 @@ static void __init cm_t3517_init_rtc(void)
>> static inline void cm_t3517_init_rtc(void) {}
>> #endif
>>
>> +#if defined(CONFIG_USB_EHCI_HCD) || defined(CONFIG_USB_EHCI_HCD_MODULE)
> Hi Igor,
Hi
> Do we really need to make these conditional on the driver being built?
This is depends on what are we trying to achieve...
I can think of two things:
[1] Simple, clear and nice code without ifdefs
[2] Smaller kernel, a bit less resources wasted and a bit faster boot time
> The hardware is present on the board at all times, right?
Yes, it is inside the SoC, except for the usb hub.
> It should be okay to execute this code independent of whether
> the driver is built or not. The device registration can be unconditional
> and if there is no driver present, we won't probe anyway.
It is ok, but again it depends on what we are trying to achieve here.
If we want [1], then indeed we need to remove the ifdefs also inside usb-ehci.c,
so pros:
* we'll get nice and clean code,
cons:
* a bit bigger kernel,
* "floating" device without a driver,
* a bit resources wasted for this device,
* a bit slower startup time (device registration, gpio toggling, etc.).
If we want [2], then we'll have the opposite of [1] ;)
Personally, I don't mind either way... ,
but I want the code to be consistent (in this case with usb-ehci.c).
> (I see some similar logic in usb-ehci.c - probably a good idea to remove
> it as well).
...
> - Anand
>
>> +#define HSUSB1_RESET_GPIO (146)
>> +#define HSUSB2_RESET_GPIO (147)
>> +#define USB_HUB_RESET_GPIO (152)
>> +
>> +static struct ehci_hcd_omap_platform_data cm_t3517_ehci_pdata __initdata = {
>> + .port_mode[0] = EHCI_HCD_OMAP_MODE_PHY,
>> + .port_mode[1] = EHCI_HCD_OMAP_MODE_PHY,
>> + .port_mode[2] = EHCI_HCD_OMAP_MODE_UNKNOWN,
>> +
>> + .phy_reset = true,
>> + .reset_gpio_port[0] = HSUSB1_RESET_GPIO,
>> + .reset_gpio_port[1] = HSUSB2_RESET_GPIO,
>> + .reset_gpio_port[2] = -EINVAL,
>> +};
>> +
>> +static int cm_t3517_init_usbh(void)
>> +{
>> + int err;
>> +
>> + err = gpio_request(USB_HUB_RESET_GPIO, "usb hub rst");
>> + if (err) {
>> + pr_err("CM-T3517: usb hub rst gpio request failed: %d\n", err);
>> + } else {
>> + gpio_direction_output(USB_HUB_RESET_GPIO, 0);
>> + udelay(10);
>> + gpio_set_value(USB_HUB_RESET_GPIO, 1);
>> + msleep(1);
>> + }
>> +
>> + usb_ehci_init(&cm_t3517_ehci_pdata);
>> +
>> + return 0;
>> +}
>> +#else
>> +static inline int cm_t3517_init_usbh(void)
>> +{
>> + return 0;
>> +}
>> +#endif
>> +
> --
> To unsubscribe from this list: send the line "unsubscribe linux-omap" in
> the body of a message to majordomo@xxxxxxxxxxxxxxx
> More majordomo info at http://vger.kernel.org/majordomo-info.html
>
--
Regards,
Igor.
--
To unsubscribe from this list: send the line "unsubscribe linux-omap" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
