Indan Zupancic
Indan Zupancic
Did you manage to fix the timer problems in sel4test?
> > Did you manage to fix the timer problems in sel4test? > > Unfortunately no. We don't use libplatsupport so I'll need to find the bandwidth to take a...
> What would you also propose as the "obvious" initial value for aarch32? I'd prefer all 0s but that's not a legal value for aarch64. There's also saving the mpidr...
> > . If 0 is what guests currently get if the VMM doesn't change it, I would be inclined to keep it that way, for backward compatibility. > >...
> Preserving the original behavior for VCPUs that don't care probably requires reading it's initial value and saving that to use as the init value for new VCPU objects. I'd...
> My preference is to leave it 0 when the object is initialized which is inline with the multicore behavior. but I think this would be a breaking change for...
> If this is the version we want, I can have a look at it soon. It's the version that Kent and I prefer, but it does change behaviour for...
We forgot to remove the `#ifdef ENABLE_SMP_SUPPORT` from `vcpu_hw_write_reg`. I opened #1224.
The DCO check failing is bogus: I just redid @alwin-joshy's commits in a way that preserves history, so the sign-off should be in his name.
Both. The only one capable of sending multiple IPIs at once is GICv2, but even there only one IPI is sent at once because that's what `generic_ipi_send_mask()` does. But the...