x86/vMSI-X: add further checks to snoop logic msixtbl_range(), as any other MMIO ->check() handlers, may get called with other than the base address of an access - avoid the snoop logic considering those. Also avoid considering vCPU-s not blocked in the hypervisor in msixtbl_pt_register(), just to be on the safe side. Signed-off-by: Jan Beulich --- a/xen/arch/x86/hvm/vmsi.c +++ b/xen/arch/x86/hvm/vmsi.c @@ -349,7 +349,7 @@ static int msixtbl_range(struct vcpu *v, { const ioreq_t *r = &v->arch.hvm_vcpu.hvm_io.io_req; - if ( r->state != STATE_IOREQ_READY ) + if ( r->state != STATE_IOREQ_READY || r->addr != addr ) return 0; ASSERT(r->type == IOREQ_TYPE_COPY); if ( r->dir == IOREQ_WRITE && r->size == 4 && !r->data_is_ptr @@ -457,7 +457,8 @@ out: for_each_vcpu ( d, v ) { - if ( v->arch.hvm_vcpu.hvm_io.msix_snoop_address == + if ( (v->pause_flags & VPF_blocked_in_xen) && + v->arch.hvm_vcpu.hvm_io.msix_snoop_address == (gtable + msi_desc->msi_attrib.entry_nr * PCI_MSIX_ENTRY_SIZE + PCI_MSIX_ENTRY_VECTOR_CTRL_OFFSET) )