| Message ID | 1390218843-15196-2-git-send-email-borntraeger@de.ibm.com (mailing list archive) |
|---|---|
| State | New, archived |
| Headers | show |
On Mon, 20 Jan 2014 12:54:03 +0100 Christian Borntraeger <borntraeger@de.ibm.com> wrote: > Seems that commit 210b1607012cc9034841a393e0591b2c86d9e26c > (KVM: s390: Removed SIE_INTERCEPT_UCONTROL) lost a hunk when we > reworked our patch queue to rework the async_fp code. We now > ignore faults on the sie instruction (guest accesses non-existing > memory) instead of sending a fault into the guest. This leads to > hang situations with the old virtio transport that checks for > descriptor memory after guest memory. Instead of bailing out this > code now goes wild... > Lets re-add the check. > > Signed-off-by: Christian Borntraeger <borntraeger@de.ibm.com> Acked-by: Cornelia Huck <cornelia.huck@de.ibm.com> > --- > arch/s390/kvm/kvm-s390.c | 4 ++++ > 1 file changed, 4 insertions(+) > > diff --git a/arch/s390/kvm/kvm-s390.c b/arch/s390/kvm/kvm-s390.c > index 1bb1dda..7635c00 100644 > --- a/arch/s390/kvm/kvm-s390.c > +++ b/arch/s390/kvm/kvm-s390.c > @@ -738,6 +738,10 @@ static int vcpu_post_run(struct kvm_vcpu *vcpu, int exit_reason) > current->thread.gmap_addr; > vcpu->run->s390_ucontrol.pgm_code = 0x10; > rc = -EREMOTE; > + } else { > + VCPU_EVENT(vcpu, 3, "%s", "fault in sie instruction"); > + trace_kvm_s390_sie_fault(vcpu); > + rc = kvm_s390_inject_program_int(vcpu, PGM_ADDRESSING); > } > > memcpy(&vcpu->run->s.regs.gprs[14], &vcpu->arch.sie_block->gg14, 16); -- To unsubscribe from this list: send the line "unsubscribe kvm" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html
diff --git a/arch/s390/kvm/kvm-s390.c b/arch/s390/kvm/kvm-s390.c index 1bb1dda..7635c00 100644 --- a/arch/s390/kvm/kvm-s390.c +++ b/arch/s390/kvm/kvm-s390.c @@ -738,6 +738,10 @@ static int vcpu_post_run(struct kvm_vcpu *vcpu, int exit_reason) current->thread.gmap_addr; vcpu->run->s390_ucontrol.pgm_code = 0x10; rc = -EREMOTE; + } else { + VCPU_EVENT(vcpu, 3, "%s", "fault in sie instruction"); + trace_kvm_s390_sie_fault(vcpu); + rc = kvm_s390_inject_program_int(vcpu, PGM_ADDRESSING); } memcpy(&vcpu->run->s.regs.gprs[14], &vcpu->arch.sie_block->gg14, 16);
Seems that commit 210b1607012cc9034841a393e0591b2c86d9e26c (KVM: s390: Removed SIE_INTERCEPT_UCONTROL) lost a hunk when we reworked our patch queue to rework the async_fp code. We now ignore faults on the sie instruction (guest accesses non-existing memory) instead of sending a fault into the guest. This leads to hang situations with the old virtio transport that checks for descriptor memory after guest memory. Instead of bailing out this code now goes wild... Lets re-add the check. Signed-off-by: Christian Borntraeger <borntraeger@de.ibm.com> --- arch/s390/kvm/kvm-s390.c | 4 ++++ 1 file changed, 4 insertions(+)