On 6/18/2025 8:53 PM, Xin Li wrote:

On 6/12/2025 2:48 PM, Sohil Mehta wrote:

quoted

From: Sean Christopherson <seanjc@google.com>

Provide separate wrappers for forwarding IRQs vs NMIs from KVM in
anticipation of adding support for NMI source reporting, which will add
an NMI-only parameter, i.e. will further pollute the current API with a
param that is a hardcoded for one of the two call sites.

Opportunistically tag the non-FRED NMI wrapper __always_inline, as the
compiler could theoretically generate a function call and trigger and a

                                                                     ^
Nit, looks to me that this is an extra "and".

Ah! Will remove.

quoted

(completely benign) "leaving noinstr" warning.

Signed-off-by: Sean Christopherson <seanjc@google.com>
Signed-off-by: Sohil Mehta <redacted>

Reviewed-by: Xin Li (Intel) <xin@zytor.com>

Thanks