On Mon, Aug 29, 2011 at 02:08:03PM +0530, ABRAHAM, KISHON VIJAY wrote: > On Mon, Aug 29, 2011 at 1:40 PM, Sebastian Andrzej Siewior > <sebastian@xxxxxxxxxxxxx> wrote: > > On Mon, Aug 29, 2011 at 12:05:40PM +0530, ABRAHAM, KISHON VIJAY wrote: > >> > - ret = usb_add_hcd(hcd, pdev->resource[1].start, > >> > - IRQF_DISABLED | IRQF_SHARED); > >> > >> here the irqflag is set to IRQF_DISABLED | IRQF_SHARED. But your > >> generic ehci_hcd_plat_add > >> sets it to IRQF_SHARED always. > > > > So? The former is defined to 0. The removal of IRQF_DISABLED is on my todo > > list. > > I see IRQF_DISABLED defined to 0x20 in my setup. Okay, it is defined to 0x20 but it is ignored by the core code. > Even then your generic usb_add_hcd shouldn't always set irqflag to > IRQF_SHARED. Why not IRQF_SHARED? What is wrong with IRQF_SHARED? We determine the source of the IRQ thanks to ehci spec. > With irq chaining for remote wakeup (still this is wip > for OMAP4), the irqflag should have IRQF_NO_SUSPEND set. We should > have a way to set irqflag to different value and this value should be > set based on the platform. So you want a way to set IRQF_NO_SUSPEND for remote wakeup? I think this could be arranged. Sebastian -- 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
