[PATCH AUTOSEL 4.9 14/20] KVM: PPC: Book3S: Use new mutex to synchronize access to rtas token list

[PATCH AUTOSEL 4.9 14/20] KVM: PPC: Book3S: Use new mutex to synchronize access to rtas token list

[Sasha Levin]

From: Paul Mackerras <paulus@ozlabs.org>

[ Upstream commit 1659e27d2bc1ef47b6d031abe01b467f18cb72d9 ]

Currently the Book 3S KVM code uses kvm->lock to synchronize access
to the kvm->arch.rtas_tokens list.  Because this list is scanned
inside kvmppc_rtas_hcall(), which is called with the vcpu mutex held,
taking kvm->lock cause a lock inversion problem, which could lead to
a deadlock.

To fix this, we add a new mutex, kvm->arch.rtas_token_lock, which nests
inside the vcpu mutexes, and use that instead of kvm->lock when
accessing the rtas token list.

This removes the lockdep_assert_held() in kvmppc_rtas_tokens_free().
At this point we don't hold the new mutex, but that is OK because
kvmppc_rtas_tokens_free() is only called when the whole VM is being
destroyed, and at that point nothing can be looking up a token in
the list.

Signed-off-by: Paul Mackerras <paulus@ozlabs.org>
Signed-off-by: Sasha Levin <sashal@kernel.org>
---
 arch/powerpc/include/asm/kvm_host.h |  1 +
 arch/powerpc/kvm/book3s.c           |  1 +
 arch/powerpc/kvm/book3s_rtas.c      | 14 ++++++--------
 3 files changed, 8 insertions(+), 8 deletions(-)

diff --git a/arch/powerpc/include/asm/kvm_host.h b/arch/powerpc/include/asm/kvm_host.h
index 5e12e19940e2..defa553fe823 100644
--- a/arch/powerpc/include/asm/kvm_host.h
+++ b/arch/powerpc/include/asm/kvm_host.h
@@ -271,6 +271,7 @@ struct kvm_arch {
 #ifdef CONFIG_PPC_BOOK3S_64
 	struct list_head spapr_tce_tables;
 	struct list_head rtas_tokens;
+	struct mutex rtas_token_lock;
 	DECLARE_BITMAP(enabled_hcalls, MAX_HCALL_OPCODE/4 + 1);
 #endif
 #ifdef CONFIG_KVM_MPIC
diff --git a/arch/powerpc/kvm/book3s.c b/arch/powerpc/kvm/book3s.c
index b6952dd23152..73c3c127d858 100644
--- a/arch/powerpc/kvm/book3s.c
+++ b/arch/powerpc/kvm/book3s.c
@@ -811,6 +811,7 @@ int kvmppc_core_init_vm(struct kvm *kvm)
 #ifdef CONFIG_PPC64
 	INIT_LIST_HEAD_RCU(&kvm->arch.spapr_tce_tables);
 	INIT_LIST_HEAD(&kvm->arch.rtas_tokens);
+	mutex_init(&kvm->arch.rtas_token_lock);
 #endif
 
 	return kvm->arch.kvm_ops->init_vm(kvm);
diff --git a/arch/powerpc/kvm/book3s_rtas.c b/arch/powerpc/kvm/book3s_rtas.c
index ef27fbd5d9c5..b1b2273d1f6d 100644
--- a/arch/powerpc/kvm/book3s_rtas.c
+++ b/arch/powerpc/kvm/book3s_rtas.c
@@ -133,7 +133,7 @@ static int rtas_token_undefine(struct kvm *kvm, char *name)
 {
 	struct rtas_token_definition *d, *tmp;
 
-	lockdep_assert_held(&kvm->lock);
+	lockdep_assert_held(&kvm->arch.rtas_token_lock);
 
 	list_for_each_entry_safe(d, tmp, &kvm->arch.rtas_tokens, list) {
 		if (rtas_name_matches(d->handler->name, name)) {
@@ -154,7 +154,7 @@ static int rtas_token_define(struct kvm *kvm, char *name, u64 token)
 	bool found;
 	int i;
 
-	lockdep_assert_held(&kvm->lock);
+	lockdep_assert_held(&kvm->arch.rtas_token_lock);
 
 	list_for_each_entry(d, &kvm->arch.rtas_tokens, list) {
 		if (d->token == token)
@@ -193,14 +193,14 @@ int kvm_vm_ioctl_rtas_define_token(struct kvm *kvm, void __user *argp)
 	if (copy_from_user(&args, argp, sizeof(args)))
 		return -EFAULT;
 
-	mutex_lock(&kvm->lock);
+	mutex_lock(&kvm->arch.rtas_token_lock);
 
 	if (args.token)
 		rc = rtas_token_define(kvm, args.name, args.token);
 	else
 		rc = rtas_token_undefine(kvm, args.name);
 
-	mutex_unlock(&kvm->lock);
+	mutex_unlock(&kvm->arch.rtas_token_lock);
 
 	return rc;
 }
@@ -232,7 +232,7 @@ int kvmppc_rtas_hcall(struct kvm_vcpu *vcpu)
 	orig_rets = args.rets;
 	args.rets = &args.args[be32_to_cpu(args.nargs)];
 
-	mutex_lock(&vcpu->kvm->lock);
+	mutex_lock(&vcpu->kvm->arch.rtas_token_lock);
 
 	rc = -ENOENT;
 	list_for_each_entry(d, &vcpu->kvm->arch.rtas_tokens, list) {
@@ -243,7 +243,7 @@ int kvmppc_rtas_hcall(struct kvm_vcpu *vcpu)
 		}
 	}
 
-	mutex_unlock(&vcpu->kvm->lock);
+	mutex_unlock(&vcpu->kvm->arch.rtas_token_lock);
 
 	if (rc == 0) {
 		args.rets = orig_rets;
@@ -269,8 +269,6 @@ void kvmppc_rtas_tokens_free(struct kvm *kvm)
 {
 	struct rtas_token_definition *d, *tmp;
 
-	lockdep_assert_held(&kvm->lock);
-
 	list_for_each_entry_safe(d, tmp, &kvm->arch.rtas_tokens, list) {
 		list_del(&d->list);
 		kfree(d);
-- 
2.20.1

[PATCH AUTOSEL 4.9 15/20] KVM: PPC: Book3S HV: Don't take kvm->lock around kvm_for_each_vcpu

[Sasha Levin]

From: Paul Mackerras <paulus@ozlabs.org>

[ Upstream commit 5a3f49364c3ffa1107bd88f8292406e98c5d206c ]

Currently the HV KVM code takes the kvm->lock around calls to
kvm_for_each_vcpu() and kvm_get_vcpu_by_id() (which can call
kvm_for_each_vcpu() internally).  However, that leads to a lock
order inversion problem, because these are called in contexts where
the vcpu mutex is held, but the vcpu mutexes nest within kvm->lock
according to Documentation/virtual/kvm/locking.txt.  Hence there
is a possibility of deadlock.

To fix this, we simply don't take the kvm->lock mutex around these
calls.  This is safe because the implementations of kvm_for_each_vcpu()
and kvm_get_vcpu_by_id() have been designed to be able to be called
locklessly.

Signed-off-by: Paul Mackerras <paulus@ozlabs.org>
Reviewed-by: Cédric Le Goater <clg@kaod.org>
Signed-off-by: Paul Mackerras <paulus@ozlabs.org>
Signed-off-by: Sasha Levin <sashal@kernel.org>
---
 arch/powerpc/kvm/book3s_hv.c | 9 +--------
 1 file changed, 1 insertion(+), 8 deletions(-)

diff --git a/arch/powerpc/kvm/book3s_hv.c b/arch/powerpc/kvm/book3s_hv.c
index 0a2b247dbc6b..e840f943cd2c 100644
--- a/arch/powerpc/kvm/book3s_hv.c
+++ b/arch/powerpc/kvm/book3s_hv.c
@@ -374,12 +374,7 @@ static void kvmppc_dump_regs(struct kvm_vcpu *vcpu)
 
 static struct kvm_vcpu *kvmppc_find_vcpu(struct kvm *kvm, int id)
 {
-	struct kvm_vcpu *ret;
-
-	mutex_lock(&kvm->lock);
-	ret = kvm_get_vcpu_by_id(kvm, id);
-	mutex_unlock(&kvm->lock);
-	return ret;
+	return kvm_get_vcpu_by_id(kvm, id);
 }
 
 static void init_vpa(struct kvm_vcpu *vcpu, struct lppaca *vpa)
@@ -1098,7 +1093,6 @@ static void kvmppc_set_lpcr(struct kvm_vcpu *vcpu, u64 new_lpcr,
 	struct kvmppc_vcore *vc = vcpu->arch.vcore;
 	u64 mask;
 
-	mutex_lock(&kvm->lock);
 	spin_lock(&vc->lock);
 	/*
 	 * If ILE (interrupt little-endian) has changed, update the
@@ -1132,7 +1126,6 @@ static void kvmppc_set_lpcr(struct kvm_vcpu *vcpu, u64 new_lpcr,
 		mask &= 0xFFFFFFFF;
 	vc->lpcr = (vc->lpcr & ~mask) | (new_lpcr & mask);
 	spin_unlock(&vc->lock);
-	mutex_unlock(&kvm->lock);
 }
 
 static int kvmppc_get_one_reg_hv(struct kvm_vcpu *vcpu, u64 id,
-- 
2.20.1