On November 11, 2022 6:23:13 AM PST, Peter Zijlstra <peterz@xxxxxxxxxxxxx> wrote: >On Fri, Nov 11, 2022 at 01:48:26PM +0100, Paolo Bonzini wrote: >> On 11/11/22 13:19, Peter Zijlstra wrote: >> > On Fri, Nov 11, 2022 at 01:04:27PM +0100, Paolo Bonzini wrote: >> > > On Intel you can optionally make it hold onto IRQs, but NMIs are always >> > > eaten by the VMEXIT and have to be reinjected manually. >> > >> > That 'optionally' thing worries me -- as in, KVM is currently >> > opting-out? >> >> Yes, because "If the “process posted interrupts” VM-execution control is 1, >> the “acknowledge interrupt on exit” VM-exit control is 1" (SDM 26.2.1.1, >> checks on VM-Execution Control Fields). Ipse dixit. Posted interrupts are >> available and used on all processors since I think Ivy Bridge. > >(imagine the non-coc compliant reaction here) > >So instead of fixing it, they made it worse :-( > >And now FRED is arguably making it worse again, and people wonder why I >hate virt... I object to saying that FRED is making it worse. Xin's patchset gets rid of low-level assembly magic across the board and at least makes it obvious what the code actually *does*, right now, as opposed to being buried in highly questionable assembly. It would also, regardless, be good to narrow down the set of possible events that may have to be reinjected to the absolute minimum, *and* document that in the code. That being said, if there are better ways of doing it, we should, and you are certainly right that we may not have properly dug into if this code is even exercised on platforms which will have FRED.
