| Message ID | 20190827013540.GC16075@blackberry (mailing list archive) |
|---|---|
| State | New, archived |
| Headers | show |
| Series | KVM: PPC: Book3S HV: Don't lose pending doorbell request on migration on P9 | expand |
On 27/08/2019 11:35, Paul Mackerras wrote: > On POWER9, when userspace reads the value of the DPDES register on a > vCPU, it is possible for 0 to be returned although there is a doorbell > interrupt pending for the vCPU. This can lead to a doorbell interrupt > being lost across migration. If the guest kernel uses doorbell > interrupts for IPIs, then it could malfunction because of the lost > interrupt. > > This happens because a newly-generated doorbell interrupt is signalled > by setting vcpu->arch.doorbell_request to 1; the DPDES value in > vcpu->arch.vcore->dpdes is not updated, because it can only be updated > when holding the vcpu mutex, in order to avoid races. > > To fix this, we OR in vcpu->arch.doorbell_request when reading the > DPDES value. > > Cc: stable@vger.kernel.org # v4.13+ > Fixes: 579006944e0d ("KVM: PPC: Book3S HV: Virtualize doorbell facility on POWER9") > Signed-off-by: Paul Mackerras <paulus@ozlabs.org> Tested-by: Alexey Kardashevskiy <aik@ozlabs.ru> > --- > arch/powerpc/kvm/book3s_hv.c | 9 ++++++++- > 1 file changed, 8 insertions(+), 1 deletion(-) > > diff --git a/arch/powerpc/kvm/book3s_hv.c b/arch/powerpc/kvm/book3s_hv.c > index ca6c6ec..88c42e7 100644 > --- a/arch/powerpc/kvm/book3s_hv.c > +++ b/arch/powerpc/kvm/book3s_hv.c > @@ -1678,7 +1678,14 @@ static int kvmppc_get_one_reg_hv(struct kvm_vcpu *vcpu, u64 id, > *val = get_reg_val(id, vcpu->arch.pspb); > break; > case KVM_REG_PPC_DPDES: > - *val = get_reg_val(id, vcpu->arch.vcore->dpdes); > + /* > + * On POWER9, where we are emulating msgsndp etc., > + * we return 1 bit for each vcpu, which can come from > + * either vcore->dpdes or doorbell_request. > + * On POWER8, doorbell_request is 0. > + */ > + *val = get_reg_val(id, vcpu->arch.vcore->dpdes | > + vcpu->arch.doorbell_request); > break; > case KVM_REG_PPC_VTB: > *val = get_reg_val(id, vcpu->arch.vcore->vtb); >
diff --git a/arch/powerpc/kvm/book3s_hv.c b/arch/powerpc/kvm/book3s_hv.c index ca6c6ec..88c42e7 100644 --- a/arch/powerpc/kvm/book3s_hv.c +++ b/arch/powerpc/kvm/book3s_hv.c @@ -1678,7 +1678,14 @@ static int kvmppc_get_one_reg_hv(struct kvm_vcpu *vcpu, u64 id, *val = get_reg_val(id, vcpu->arch.pspb); break; case KVM_REG_PPC_DPDES: - *val = get_reg_val(id, vcpu->arch.vcore->dpdes); + /* + * On POWER9, where we are emulating msgsndp etc., + * we return 1 bit for each vcpu, which can come from + * either vcore->dpdes or doorbell_request. + * On POWER8, doorbell_request is 0. + */ + *val = get_reg_val(id, vcpu->arch.vcore->dpdes | + vcpu->arch.doorbell_request); break; case KVM_REG_PPC_VTB: *val = get_reg_val(id, vcpu->arch.vcore->vtb);
On POWER9, when userspace reads the value of the DPDES register on a vCPU, it is possible for 0 to be returned although there is a doorbell interrupt pending for the vCPU. This can lead to a doorbell interrupt being lost across migration. If the guest kernel uses doorbell interrupts for IPIs, then it could malfunction because of the lost interrupt. This happens because a newly-generated doorbell interrupt is signalled by setting vcpu->arch.doorbell_request to 1; the DPDES value in vcpu->arch.vcore->dpdes is not updated, because it can only be updated when holding the vcpu mutex, in order to avoid races. To fix this, we OR in vcpu->arch.doorbell_request when reading the DPDES value. Cc: stable@vger.kernel.org # v4.13+ Fixes: 579006944e0d ("KVM: PPC: Book3S HV: Virtualize doorbell facility on POWER9") Signed-off-by: Paul Mackerras <paulus@ozlabs.org> --- arch/powerpc/kvm/book3s_hv.c | 9 ++++++++- 1 file changed, 8 insertions(+), 1 deletion(-)