Re: [RFC PATCH] PCI: remove access to pci_[enable|disable]_msi()for drivers - take 2

[Roland Dreier]

    Andi> I think we should have the right (default MSI) API by
    Andi> default.  If we wait 5 years we end up with lots of mess in
    Andi> the drivers.

I agree that we should get the API correct as soon as we can.  However
I would argue that one property of the correct API would be that
changing the default from MSI turned off to MSI turned on should not
affect the API.

I like Jeff's suggestion of rolling more common code into
pci_enable_device().  However I'm not sure if this fixes the MSI mess
completely -- there may be some devices where the driver wants to test
interrupt generation and disable MSI if it fails.  However this may be
enough of a special case that we can just add a flag to those
particular drivers so that users can manually control the use of MSI.

 - R.

-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/