Re: PCIE

[Roland Dreier <rdreier@cisco.com>]

 > Presumably we need something like IRQF_MSI which can be set as
 > appropriate depending on the architecture?

Unfortunately I don't think it's that simple -- drivers would have to
do something like

	if (IRQF_MSI == IRQF_SHARED) {
		// lose MSI optimizations, do an MMIO read, etc.
	} else...

As I said I think that if we're running on a system where MSI
interrupts might be shared, we should just have pci_enable_msi() fail
so drivers fall back to their usual interrupt handler.
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/