On 11/01/2023 18:06, David Woodhouse wrote:
From: David Woodhouse <dwmw@xxxxxxxxxxxx>
The kvm_xen_update_runstate_guest() function can be called when the vCPU
is being scheduled out, from a preempt notifier. It *opportunistically*
updates the runstate area in the guest memory, if the gfn_to_pfn_cache
which caches the appropriate address is still valid.
If there is *contention* when it attempts to obtain gpc->lock, then
locking inside the priority inheritance checks may cause a deadlock.
Lockdep reports:
[13890.148997] Chain exists of:
&gpc->lock --> &p->pi_lock --> &rq->__lock
[13890.149002] Possible unsafe locking scenario:
[13890.149003] CPU0 CPU1
[13890.149004] ---- ----
[13890.149005] lock(&rq->__lock);
[13890.149007] lock(&p->pi_lock);
[13890.149009] lock(&rq->__lock);
[13890.149011] lock(&gpc->lock);
[13890.149013]
*** DEADLOCK ***
In the general case, if there's contention for a read lock on gpc->lock,
that's going to be because something else is either invalidating or
revalidating the cache. Either way, we've raced with seeing it in an
invalid state, in which case we would have aborted the opportunistic
update anyway.
So in the 'atomic' case when called from the preempt notifier, just
switch to using read_trylock() and avoid the PI handling altogether.
Signed-off-by: David Woodhouse <dwmw@xxxxxxxxxxxx>
---
arch/x86/kvm/xen.c | 19 +++++++++++++++++--
1 file changed, 17 insertions(+), 2 deletions(-)
Reviewed-by: Paul Durrant <paul@xxxxxxx>
... with one suggestion below.
diff --git a/arch/x86/kvm/xen.c b/arch/x86/kvm/xen.c
index 07e61cc9881e..809b82bdb9c3 100644
--- a/arch/x86/kvm/xen.c
+++ b/arch/x86/kvm/xen.c
@@ -272,7 +272,15 @@ static void kvm_xen_update_runstate_guest(struct kvm_vcpu *v, bool atomic)
* Attempt to obtain the GPC lock on *both* (if there are two)
* gfn_to_pfn caches that cover the region.
*/
- read_lock_irqsave(&gpc1->lock, flags);
+ if (atomic) {
+ local_irq_save(flags);
+ if (!read_trylock(&gpc1->lock)) {
+ local_irq_restore(flags);
+ return;
+ }
+ } else {
+ read_lock_irqsave(&gpc1->lock, flags);
+ }
while (!kvm_gpc_check(gpc1, user_len1)) {
read_unlock_irqrestore(&gpc1->lock, flags);
@@ -309,7 +317,14 @@ static void kvm_xen_update_runstate_guest(struct kvm_vcpu *v, bool atomic)
* gpc1 lock to make lockdep shut up about it.
*/
lock_set_subclass(&gpc1->lock.dep_map, 1, _THIS_IP_);
- read_lock(&gpc2->lock);
+ if (atomic) {
+ if (!read_trylock(&gpc2->lock)) {
You could avoid the nesting in this case with:
if (atomic && !read_trylock(&gpc2->lock))
+ read_unlock_irqrestore(&gpc1->lock, flags);
+ return;
+ }
+ } else {
+ read_lock(&gpc2->lock);
+ }
if (!kvm_gpc_check(gpc2, user_len2)) {
read_unlock(&gpc2->lock);
