On 21/05/20 16:59, Vitaly Kuznetsov wrote: >> However, interrupts for 'page ready' do have a bunch of advantages (more >> control on what can be preempted by the notification, a saner check for >> new page faults which is effectively a bug fix) so it makes sense to get >> them in more quickly (probably 5.9 at this point due to the massive >> cleanups that are being done around interrupt vectors). >> > Actually, I have almost no feedback to address in v2 :-) Almost all > discussion are happening around #VE. Don't mean to rush or anything but > if the 'cleanups' are finalized I can hopefully rebase and retest very > quickly as it's only the KVM guest part which intersects with them, the > rest should be KVM-only. But 5.9 is good too) Yeah, going for 5.9 would only be due to the conflicts. Do send v2 anyway now, so that we can use a merge commit to convert the interrupt vector to the 5.8 style. Paolo
