From: David Hildenbrand <dahi@xxxxxxxxxxxxxxxxxx>
Thinking about it, I can't find a real use case where we want
to block a VCPU and not kick it out of SIE. (except if we want
to do the same in batch for multiple VCPUs - but that's a micro
optimization)
So let's simply perform the exit_sie() calls directly when setting
the other magic block bits in the SIE.
Otherwise e.g. kvm_s390_set_tod_low() still has other VCPUs running
after that call, working with a wrong epoch.
Fixes: 27406cd50c ("KVM: s390: provide functions for blocking all CPUs")
Acked-by: Christian Borntraeger <borntraeger@xxxxxxxxxx>
Signed-off-by: David Hildenbrand <dahi@xxxxxxxxxxxxxxxxxx>
Signed-off-by: Christian Borntraeger <borntraeger@xxxxxxxxxx>
---
arch/s390/kvm/kvm-s390.c | 3 ++-
1 file changed, 2 insertions(+), 1 deletion(-)
diff --git a/arch/s390/kvm/kvm-s390.c b/arch/s390/kvm/kvm-s390.c
index 6bc69ab..f37557b 100644
--- a/arch/s390/kvm/kvm-s390.c
+++ b/arch/s390/kvm/kvm-s390.c
@@ -1417,6 +1417,7 @@ int kvm_arch_vcpu_runnable(struct kvm_vcpu *vcpu)
void kvm_s390_vcpu_block(struct kvm_vcpu *vcpu)
{
atomic_set_mask(PROG_BLOCK_SIE, &vcpu->arch.sie_block->prog20);
+ exit_sie(vcpu);
}
void kvm_s390_vcpu_unblock(struct kvm_vcpu *vcpu)
@@ -1427,6 +1428,7 @@ void kvm_s390_vcpu_unblock(struct kvm_vcpu *vcpu)
static void kvm_s390_vcpu_request(struct kvm_vcpu *vcpu)
{
atomic_set_mask(PROG_REQUEST, &vcpu->arch.sie_block->prog20);
+ exit_sie(vcpu);
}
static void kvm_s390_vcpu_request_handled(struct kvm_vcpu *vcpu)
@@ -1450,7 +1452,6 @@ void kvm_s390_sync_request(int req, struct kvm_vcpu *vcpu)
{
kvm_make_request(req, vcpu);
kvm_s390_vcpu_request(vcpu);
- exit_sie(vcpu);
}
static void kvm_gmap_notifier(struct gmap *gmap, unsigned long address)
--
2.3.0
--
To unsubscribe from this list: send the line "unsubscribe kvm" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
