Re: [Xen-devel] [PATCH v3 2/5] x86: track when in #NMI context

[Jan Beulich <jbeulich@xxxxxxxx>]

On 24.02.2020 11:46, Roger Pau Monne wrote:
> Add helpers to track when running in #NMI context. This is modeled
> after the in_irq helpers.
> 
> The SDM states that no #NMI can be delivered while handling a #NMI
> until the processor has executed an iret instruction. It's possible
> however that another fault is received while handling the #NMI (a #MC
> for example), and thus the iret from that fault would allow further
> #NMIs to be injected while still processing the previous one, and
> hence an integer is needed in order to keep track of in service #NMIs.

While I agree that this needs taking care of, I'm afraid in_nmi()
is becoming ambiguous because of this - you give it the meaning
"we're handling an NMI", while one could also assume it to mean
"we're in NMI context, i.e. further NMIs are not possible". IOW
I think we want to consider using another name, despite this
getting things less nicely aligned with in_irq(). Perhaps
in_nmi_handler()?

> While there move nmi_count() into a x86 specific header and drop the
> leading underscores from __nmi_count field.

I notice you re-use the fields that I suggested to be removed by
the prior patch. I think logically they should indeed be removed
there, and a new field and a new macro be introduced here.

Jan

_______________________________________________
Xen-devel mailing list
Xen-devel@xxxxxxxxxxxxxxxxxxxx
https://lists.xenproject.org/mailman/listinfo/xen-devel