diff options
author | Dominik Dingel <dingel@linux.vnet.ibm.com> | 2014-03-25 13:47:11 +0100 |
---|---|---|
committer | Christian Borntraeger <borntraeger@de.ibm.com> | 2014-04-22 13:24:13 +0200 |
commit | b31605c12f4edda7a40ad9f86985739666298c31 (patch) | |
tree | 6dbda314904aeef619702812f7b6aab836459dde /arch/s390/kvm/priv.c | |
parent | f2061656209fb9a5d54bbb1999f0a633438504e7 (diff) |
KVM: s390: make cmma usage conditionally
When userspace reset the guest without notifying kvm, the CMMA state
of the pages might be unused, resulting in guest data corruption.
To avoid this, CMMA must be enabled only if userspace understands
the implications.
CMMA must be enabled before vCPU creation. It can't be switched off
once enabled. All subsequently created vCPUs will be enabled for
CMMA according to the CMMA state of the VM.
Signed-off-by: Dominik Dingel <dingel@linux.vnet.ibm.com>
Signed-off-by: Christian Borntraeger <borntraeger@de.ibm.com>
[remove now unnecessary calls to page_table_reset_pgste]
Diffstat (limited to 'arch/s390/kvm/priv.c')
-rw-r--r-- | arch/s390/kvm/priv.c | 2 |
1 files changed, 1 insertions, 1 deletions
diff --git a/arch/s390/kvm/priv.c b/arch/s390/kvm/priv.c index 8a63e992936..9a04d74c5fb 100644 --- a/arch/s390/kvm/priv.c +++ b/arch/s390/kvm/priv.c @@ -656,7 +656,7 @@ static int handle_essa(struct kvm_vcpu *vcpu) VCPU_EVENT(vcpu, 5, "cmma release %d pages", entries); gmap = vcpu->arch.gmap; vcpu->stat.instruction_essa++; - if (!kvm_enabled_cmma() || !vcpu->arch.sie_block->cbrlo) + if (!kvm_s390_cmma_enabled(vcpu->kvm)) return kvm_s390_inject_program_int(vcpu, PGM_OPERATION); if (vcpu->arch.sie_block->gpsw.mask & PSW_MASK_PSTATE) |