* [GIT PULL 1/1] KVM: s390: add proper locking for CMMA migration bitmap
2018-01-24 14:49 [GIT PULL 0/1] KVM: s390: another fix for 4.15 (via kvm/master) Christian Borntraeger
@ 2018-01-24 14:49 ` Christian Borntraeger
2018-01-25 14:22 ` [GIT PULL 0/1] KVM: s390: another fix for 4.15 (via kvm/master) Radim Krčmář
1 sibling, 0 replies; 3+ messages in thread
From: Christian Borntraeger @ 2018-01-24 14:49 UTC (permalink / raw)
To: Paolo Bonzini, Radim Krčmář
Cc: KVM, Cornelia Huck, Christian Borntraeger, linux-s390,
Janosch Frank, David Hildenbrand, Claudio Imbrenda
Some parts of the cmma migration bitmap is already protected
with the kvm->lock (e.g. the migration start). On the other
hand the read of the cmma bits is not protected against a
concurrent free, neither is the emulation of the ESSA instruction.
Let's extend the locking to all related ioctls by using
the slots lock for
- kvm_s390_vm_start_migration
- kvm_s390_vm_stop_migration
- kvm_s390_set_cmma_bits
- kvm_s390_get_cmma_bits
In addition to that, we use synchronize_srcu before freeing
the migration structure as all users hold kvm->srcu for read.
(e.g. the ESSA handler).
Reported-by: David Hildenbrand <david@redhat.com>
Signed-off-by: Christian Borntraeger <borntraeger@de.ibm.com>
Cc: stable@vger.kernel.org # 4.13+
Fixes: 190df4a212a7 (KVM: s390: CMMA tracking, ESSA emulation, migration mode)
Reviewed-by: Claudio Imbrenda <imbrenda@linux.vnet.ibm.com>
Reviewed-by: David Hildenbrand <david@redhat.com>
Reviewed-by: Cornelia Huck <cohuck@redhat.com>
---
arch/s390/kvm/kvm-s390.c | 18 +++++++++++-------
1 file changed, 11 insertions(+), 7 deletions(-)
diff --git a/arch/s390/kvm/kvm-s390.c b/arch/s390/kvm/kvm-s390.c
index abcd24fdde3fc..52880e980a336 100644
--- a/arch/s390/kvm/kvm-s390.c
+++ b/arch/s390/kvm/kvm-s390.c
@@ -766,7 +766,7 @@ static void kvm_s390_sync_request_broadcast(struct kvm *kvm, int req)
/*
* Must be called with kvm->srcu held to avoid races on memslots, and with
- * kvm->lock to avoid races with ourselves and kvm_s390_vm_stop_migration.
+ * kvm->slots_lock to avoid races with ourselves and kvm_s390_vm_stop_migration.
*/
static int kvm_s390_vm_start_migration(struct kvm *kvm)
{
@@ -822,7 +822,7 @@ static int kvm_s390_vm_start_migration(struct kvm *kvm)
}
/*
- * Must be called with kvm->lock to avoid races with ourselves and
+ * Must be called with kvm->slots_lock to avoid races with ourselves and
* kvm_s390_vm_start_migration.
*/
static int kvm_s390_vm_stop_migration(struct kvm *kvm)
@@ -837,6 +837,8 @@ static int kvm_s390_vm_stop_migration(struct kvm *kvm)
if (kvm->arch.use_cmma) {
kvm_s390_sync_request_broadcast(kvm, KVM_REQ_STOP_MIGRATION);
+ /* We have to wait for the essa emulation to finish */
+ synchronize_srcu(&kvm->srcu);
vfree(mgs->pgste_bitmap);
}
kfree(mgs);
@@ -846,14 +848,12 @@ static int kvm_s390_vm_stop_migration(struct kvm *kvm)
static int kvm_s390_vm_set_migration(struct kvm *kvm,
struct kvm_device_attr *attr)
{
- int idx, res = -ENXIO;
+ int res = -ENXIO;
- mutex_lock(&kvm->lock);
+ mutex_lock(&kvm->slots_lock);
switch (attr->attr) {
case KVM_S390_VM_MIGRATION_START:
- idx = srcu_read_lock(&kvm->srcu);
res = kvm_s390_vm_start_migration(kvm);
- srcu_read_unlock(&kvm->srcu, idx);
break;
case KVM_S390_VM_MIGRATION_STOP:
res = kvm_s390_vm_stop_migration(kvm);
@@ -861,7 +861,7 @@ static int kvm_s390_vm_set_migration(struct kvm *kvm,
default:
break;
}
- mutex_unlock(&kvm->lock);
+ mutex_unlock(&kvm->slots_lock);
return res;
}
@@ -1751,7 +1751,9 @@ long kvm_arch_vm_ioctl(struct file *filp,
r = -EFAULT;
if (copy_from_user(&args, argp, sizeof(args)))
break;
+ mutex_lock(&kvm->slots_lock);
r = kvm_s390_get_cmma_bits(kvm, &args);
+ mutex_unlock(&kvm->slots_lock);
if (!r) {
r = copy_to_user(argp, &args, sizeof(args));
if (r)
@@ -1765,7 +1767,9 @@ long kvm_arch_vm_ioctl(struct file *filp,
r = -EFAULT;
if (copy_from_user(&args, argp, sizeof(args)))
break;
+ mutex_lock(&kvm->slots_lock);
r = kvm_s390_set_cmma_bits(kvm, &args);
+ mutex_unlock(&kvm->slots_lock);
break;
}
default:
--
2.13.4
^ permalink raw reply related [flat|nested] 3+ messages in thread* Re: [GIT PULL 0/1] KVM: s390: another fix for 4.15 (via kvm/master)
2018-01-24 14:49 [GIT PULL 0/1] KVM: s390: another fix for 4.15 (via kvm/master) Christian Borntraeger
2018-01-24 14:49 ` [GIT PULL 1/1] KVM: s390: add proper locking for CMMA migration bitmap Christian Borntraeger
@ 2018-01-25 14:22 ` Radim Krčmář
1 sibling, 0 replies; 3+ messages in thread
From: Radim Krčmář @ 2018-01-25 14:22 UTC (permalink / raw)
To: Christian Borntraeger
Cc: Paolo Bonzini, KVM, Cornelia Huck, linux-s390, Janosch Frank,
David Hildenbrand, Claudio Imbrenda
2018-01-24 15:49+0100, Christian Borntraeger:
> Paolo, Radim,
>
> one last fix for 4.15. (if it is too late for 4.15 I can send it
> for the next merge window together with the other pending patches, let
> me know).
It is a small fix, so no problem for 4.15. Pulled, thanks.
^ permalink raw reply [flat|nested] 3+ messages in thread