On Fri, Oct 04, 2013 at 09:31:49AM +0100, David Laight wrote:

quoted

Mmmm.. I am not sure I am getting it. Could you please rephrase?

One possibility is for drivers than can use a lot of interrupts to
request a minimum number initially and then request the additional
ones much later on.
That would make it less likely that none will be available for
devices/drivers that need them but are initialised later.

It sounds as a whole new topic for me. Isn't it?

Anyway, what prevents the above from being done with pci_enable_msix(nvec1) -
pci_disable_msix() - pci_enable_msix(nvec2) where nvec1 < nvec2?

	David

-- 
Regards,
Alexander Gordeev
agordeev@redhat.com