On 06/08/15 11:05, Julien Grall wrote: >> But Ian suggested to export gicv3_host_irq_end instead of calling >> gicv3_eoi_irq() > > And Ian said "Exposing those two gicv3 functions is a bit unfortunate, > but I think it will do for now.". > > Which means he was opposed to the previous solution.
*he wasn't opposed. > > Anyway, I think it's a fair trade compare to the overhead you had for > everytime you received an IRQ used by Xen. This is more true given that > we don't even support LPI for Xen... > > Regards, > -- Julien Grall _______________________________________________ Xen-devel mailing list Xen-devel@lists.xen.org http://lists.xen.org/xen-devel
