Re: [PATCH v5 1/5] arm64: Add framework to turn IPI as NMI

[Sumit Garg]

On Mon, 19 Oct 2020 at 17:26, Marc Zyngier <maz@xxxxxxxxxx> wrote:
>
> On 2020-10-14 12:12, Sumit Garg wrote:
> > Introduce framework to turn an IPI as NMI using pseudo NMIs. In case a
> > particular platform doesn't support pseudo NMIs, then request IPI as a
> > regular IRQ.
> >
> > The main motivation for this feature is to have an IPI that can be
> > leveraged to invoke NMI functions on other CPUs. And current
> > prospective
> > users are NMI backtrace and KGDB CPUs round-up whose support is added
> > via future patches.
> >
> > Signed-off-by: Sumit Garg <sumit.garg@xxxxxxxxxx>
> > ---
> >  arch/arm64/include/asm/nmi.h | 16 +++++++++
> >  arch/arm64/kernel/Makefile   |  2 +-
> >  arch/arm64/kernel/ipi_nmi.c  | 77
> > ++++++++++++++++++++++++++++++++++++++++++++
> >  3 files changed, 94 insertions(+), 1 deletion(-)
> >  create mode 100644 arch/arm64/include/asm/nmi.h
> >  create mode 100644 arch/arm64/kernel/ipi_nmi.c
>
> [...]
>
> > +     irq_set_status_flags(ipi, IRQ_HIDDEN);
>
> Another thing is this. Why are you hiding this from /proc/interrupts?
> The only reason the other IPIs are hidden is that displaying them as
> "normal" interrupts would be a change in userspace ABI.
>
> In your case, this is something new that can perfectly appear as
> a standard interrupt (and I don't see how you'd display the
> statistics otherwise).

Makes sense. I will remove this flag for this IPI so that it can be
displayed as a standard interrupt.

-Sumit

>
>          M.
> --
> Jazz is not dead. It just smells funny...