On Thu, Feb 18, 2021 at 11:42:01PM +0900, Hector Martin wrote: > On 18/02/2021 23.22, Mark Rutland wrote: > > I think that for consistency we always want to keep IRQ and FIQ in-sync, > > even when using GIC priorities. So when handling a pseudo-NMI we should > > unmask DAIF.DA and leave DAIF.IF masked. > > In that case there's one more, in daifflags.h:local_daif_restore(): > > /* > * If interrupts are disabled but we can take > * asynchronous errors, we can take NMIs > */ > flags &= PSR_I_BIT; > pmr = GIC_PRIO_IRQOFF;
Good spot, yes! I did a quick scan with `git grep 'PSR_[IF]_BIT' -- arch/arm64`, and AFAICT that's the last one. > > > And a minor related one: should init_gic_priority_masking() WARN if FIQ is > > > masked too? This probably goes with the above. > > > > I think it should, yes. > > Done for v3 then. Thanks! Cool! Mark.
