Re: [Qemu-devel] [PATCH/RFC 4/5] s390x/kvm: test whether a cpu is STOPPED when checking "has_work"

[Paolo Bonzini <pbonzini@redhat.com>]

Il 28/07/2014 17:03, David Hildenbrand ha scritto:
> Well the difference is, that a STOPPED vcpu can be woken up by non-interrupt
> like things (SIGP START) AND a special interrupt (SIGP RESTART - which is like
> a "SIPI"++ as it performs a psw exchange - "NMI"). So we basically have two
> paths that can lead to a state change.  All interrupt bits may be in any
> combination (SIGP RESTART interrupts can't be masked out, nor can SIGP START be
> denied).
> 
> The other thing may be that on s390, each vcpu (including itself) can put
> another vcpu into the STOPPED state - I assume that this is different for x86 "
> INIT_RECEIVED". For this reason we have to watch out for bad race conditions
> (e.g. multiple vcpus working on another vcpu)...

You can do that in x86 by sending an INIT inter-processor interrupt.  A
SIPI is ignored if the CPU is not in INIT_RECEIVED state.

Commit 66450a21f99636af4fafac2afd33f1a40631bc3a introduced the current
implementation.

- an INIT cancels a previous SIPI;

- if both INIT and SIPI are sent, on real hardware you need to have a
few hundred microseconds between them, but KVM will reliably process
INIT before SIPI.

See commit 299018f44ac553dce3caf84df1d14c4764faa279 for an example of
the races that can happen.

Note that x86 has KVM_MP_STATE_SIPI_RECEIVED state but it is obsolete,
we go straight from KVM_MP_STATE_INIT_RECEIVED to KVM_MP_STATE_RUNNABLE.