On Fri, 3 Dec 2010, Wu, Hao wrote:
> >> +#elif defined(CONFIG_X86_MRST)
> >> +static void hcd_pci_set_platform_config(struct pci_dev *pci_dev, int enable)
> >> +{
> >> + /* Check if it is Intel Medfield platform */
> >> + if (pci_dev->vendor == PCI_VENDOR_ID_INTEL &&
> >> + pci_dev->device == 0x0829) {
> >> + if (enable) {
> >> + pci_restore_state(pci_dev);
> >> + pci_set_power_state(pci_dev, PCI_D0);
> >> + } else {
> >> + pci_save_state(pci_dev);
> >> + pci_set_power_state(pci_dev, PCI_D3hot);
> >> + }
> >> + }
> >> +}
> >
> >Why do you need this routine? Doesn't the PCI core already do this for
> >you?
>
> Hello
>
> Thanks a lot for your help on code review.
>
> we manually invoke pci_set_power_state to notify that the hardware can be powered on/off in this function mainly for runtime pm case.
> Actually it depends on Intel MID platform related pm implementation, it will not enter D3hot in runtime suspend by PCI core but required driver to call pci_set_power_state itself.
Then there's a bug in the Intel MID platform-related PM implementation.
You should fix that bug instead of introducing extra unnecessary code
somewhere else. After all, that same bug will affect other PCI drivers
on the MID platform.
Alan Stern
--
To unsubscribe from this list: send the line "unsubscribe linux-usb" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
