Import of kernel-6.12.0-55.30.1.el10_0

Makefile.rhelver

@@ -12,7 +12,7 @@ RHEL_MINOR = 0
 #
 # Use this spot to avoid future merge conflicts.
 # Do not trim this comment.
-RHEL_RELEASE = 55.29.1
+RHEL_RELEASE = 55.30.1
 
 #
 # RHEL_REBASE_NUM

arch/s390/pci/pci_bus.c

@@ -171,7 +171,6 @@ void zpci_bus_scan_busses(void)
 static bool zpci_bus_is_multifunction_root(struct zpci_dev *zdev)
 {
 	return !s390_pci_no_rid && zdev->rid_available &&
-		zpci_is_device_configured(zdev) &&
 		!zdev->vfn;
 }
 
@@ -332,6 +331,20 @@ error:
 	return rc;
 }
 
+static bool zpci_bus_is_isolated_vf(struct zpci_bus *zbus, struct zpci_dev *zdev)
+{
+	struct pci_dev *pdev;
+
+	if (!zdev->vfn)
+		return false;
+
+	pdev = zpci_iov_find_parent_pf(zbus, zdev);
+	if (!pdev)
+		return true;
+	pci_dev_put(pdev);
+	return false;
+}
+
 int zpci_bus_device_register(struct zpci_dev *zdev, struct pci_ops *ops)
 {
 	bool topo_is_tid = zdev->tid_avail;
@@ -346,6 +359,15 @@ int zpci_bus_device_register(struct zpci_dev *zdev, struct pci_ops *ops)
 
 	topo = topo_is_tid ? zdev->tid : zdev->pchid;
 	zbus = zpci_bus_get(topo, topo_is_tid);
+	/*
+	 * An isolated VF gets its own domain/bus even if there exists
+	 * a matching domain/bus already
+	 */
+	if (zbus && zpci_bus_is_isolated_vf(zbus, zdev)) {
+		zpci_bus_put(zbus);
+		zbus = NULL;
+	}
+
 	if (!zbus) {
 		zbus = zpci_bus_alloc(topo, topo_is_tid);
 		if (!zbus)

arch/s390/pci/pci_iov.c

@@ -60,18 +60,35 @@ static int zpci_iov_link_virtfn(struct pci_dev *pdev, struct pci_dev *virtfn, in
 	return 0;
 }
 
-int zpci_iov_setup_virtfn(struct zpci_bus *zbus, struct pci_dev *virtfn, int vfn)
+/**
+ * zpci_iov_find_parent_pf - Find the parent PF, if any, of the given function
+ * @zbus:	The bus that the PCI function is on, or would be added on
+ * @zdev:	The PCI function
+ *
+ * Finds the parent PF, if it exists and is configured, of the given PCI function
+ * and increments its refcount. Th PF is searched for on the provided bus so the
+ * caller has to ensure that this is the correct bus to search. This function may
+ * be used before adding the PCI function to a zbus.
+ *
+ * Return: Pointer to the struct pci_dev of the parent PF or NULL if it not
+ * found. If the function is not a VF or has no RequesterID information,
+ * NULL is returned as well.
+ */
+struct pci_dev *zpci_iov_find_parent_pf(struct zpci_bus *zbus, struct zpci_dev *zdev)
 {
-	int i, cand_devfn;
+	int i, vfid, devfn, cand_devfn;
-	struct zpci_dev *zdev;
 	struct pci_dev *pdev;
-	int vfid = vfn - 1; /* Linux' vfid's start at 0 vfn at 1*/
-	int rc = 0;
 
 	if (!zbus->multifunction)
-		return 0;
+		return NULL;
-
+	/* Non-VFs and VFs without RID available don't have a parent */
-	/* If the parent PF for the given VF is also configured in the
+	if (!zdev->vfn || !zdev->rid_available)
+		return NULL;
+	/* Linux vfid starts at 0 vfn at 1 */
+	vfid = zdev->vfn - 1;
+	devfn = zdev->rid & ZPCI_RID_MASK_DEVFN;
+	/*
+	 * If the parent PF for the given VF is also configured in the
 	 * instance, it must be on the same zbus.
 	 * We can then identify the parent PF by checking what
 	 * devfn the VF would have if it belonged to that PF using the PF's
@@ -85,15 +102,26 @@ int zpci_iov_setup_virtfn(struct zpci_bus *zbus, struct pci_dev *virtfn, int vfn
 			if (!pdev)
 				continue;
 			cand_devfn = pci_iov_virtfn_devfn(pdev, vfid);
-			if (cand_devfn == virtfn->devfn) {
+			if (cand_devfn == devfn)
-				rc = zpci_iov_link_virtfn(pdev, virtfn, vfid);
+				return pdev;
-				/* balance pci_get_slot() */
-				pci_dev_put(pdev);
-				break;
-			}
 			/* balance pci_get_slot() */
 			pci_dev_put(pdev);
 		}
 	}
+	return NULL;
+}
+
+int zpci_iov_setup_virtfn(struct zpci_bus *zbus, struct pci_dev *virtfn, int vfn)
+{
+	struct zpci_dev *zdev = to_zpci(virtfn);
+	struct pci_dev *pdev_pf;
+	int rc = 0;
+
+	pdev_pf = zpci_iov_find_parent_pf(zbus, zdev);
+	if (pdev_pf) {
+		/* Linux' vfids start at 0 while zdev->vfn starts at 1 */
+		rc = zpci_iov_link_virtfn(pdev_pf, virtfn, zdev->vfn - 1);
+		pci_dev_put(pdev_pf);
+	}
 	return rc;
 }

arch/s390/pci/pci_iov.h

@@ -17,6 +17,8 @@ void zpci_iov_map_resources(struct pci_dev *pdev);
 
 int zpci_iov_setup_virtfn(struct zpci_bus *zbus, struct pci_dev *virtfn, int vfn);
 
+struct pci_dev *zpci_iov_find_parent_pf(struct zpci_bus *zbus, struct zpci_dev *zdev);
+
 #else /* CONFIG_PCI_IOV */
 static inline void zpci_iov_remove_virtfn(struct pci_dev *pdev, int vfn) {}
 
@@ -26,5 +28,10 @@ static inline int zpci_iov_setup_virtfn(struct zpci_bus *zbus, struct pci_dev *v
 {
 	return 0;
 }
+
+static inline struct pci_dev *zpci_iov_find_parent_pf(struct zpci_bus *zbus, struct zpci_dev *zdev)
+{
+	return NULL;
+}
 #endif /* CONFIG_PCI_IOV */
 #endif /* __S390_PCI_IOV_h */

drivers/infiniband/core/iwcm.c

@@ -366,12 +366,9 @@ EXPORT_SYMBOL(iw_cm_disconnect);
 /*
  * CM_ID <-- DESTROYING
  *
- * Clean up all resources associated with the connection and release
+ * Clean up all resources associated with the connection.
- * the initial reference taken by iw_create_cm_id.
- *
- * Returns true if and only if the last cm_id_priv reference has been dropped.
  */
-static bool destroy_cm_id(struct iw_cm_id *cm_id)
+static void destroy_cm_id(struct iw_cm_id *cm_id)
 {
 	struct iwcm_id_private *cm_id_priv;
 	struct ib_qp *qp;
@@ -440,20 +437,22 @@ static bool destroy_cm_id(struct iw_cm_id *cm_id)
 		iwpm_remove_mapinfo(&cm_id->local_addr, &cm_id->m_local_addr);
 		iwpm_remove_mapping(&cm_id->local_addr, RDMA_NL_IWCM);
 	}
-
-	return iwcm_deref_id(cm_id_priv);
 }
 
 /*
- * This function is only called by the application thread and cannot
+ * Destroy cm_id. If the cm_id still has other references, wait for all
- * be called by the event thread. The function will wait for all
+ * references to be released on the cm_id and then release the initial
- * references to be released on the cm_id and then kfree the cm_id
+ * reference taken by iw_create_cm_id.
- * object.
  */
 void iw_destroy_cm_id(struct iw_cm_id *cm_id)
 {
-	if (!destroy_cm_id(cm_id))
+	struct iwcm_id_private *cm_id_priv;
+
+	cm_id_priv = container_of(cm_id, struct iwcm_id_private, id);
+	destroy_cm_id(cm_id);
+	if (refcount_read(&cm_id_priv->refcount) > 1)
 		flush_workqueue(iwcm_wq);
+	iwcm_deref_id(cm_id_priv);
 }
 EXPORT_SYMBOL(iw_destroy_cm_id);
 
@@ -1033,8 +1032,10 @@ static void cm_work_handler(struct work_struct *_work)
 
 		if (!test_bit(IWCM_F_DROP_EVENTS, &cm_id_priv->flags)) {
 			ret = process_event(cm_id_priv, &levent);
-			if (ret)
+			if (ret) {
-				WARN_ON_ONCE(destroy_cm_id(&cm_id_priv->id));
+				destroy_cm_id(&cm_id_priv->id);
+				WARN_ON_ONCE(iwcm_deref_id(cm_id_priv));
+			}
 		} else
 			pr_debug("dropping event %d\n", levent.event);
 		if (iwcm_deref_id(cm_id_priv))

fs/ext4/inode.c

@@ -998,7 +998,12 @@ int ext4_walk_page_buffers(handle_t *handle, struct inode *inode,
  */
 static int ext4_dirty_journalled_data(handle_t *handle, struct buffer_head *bh)
 {
-	folio_mark_dirty(bh->b_folio);
+	struct folio *folio = bh->b_folio;
+	struct inode *inode = folio->mapping->host;
+
+	/* only regular files have a_ops */
+	if (S_ISREG(inode->i_mode))
+		folio_mark_dirty(folio);
 	return ext4_handle_dirty_metadata(handle, NULL, bh);
 }
 

include/net/netfilter/nf_conntrack.h

@@ -308,8 +308,19 @@ static inline bool nf_ct_is_expired(const struct nf_conn *ct)
 /* use after obtaining a reference count */
 static inline bool nf_ct_should_gc(const struct nf_conn *ct)
 {
-	return nf_ct_is_expired(ct) && nf_ct_is_confirmed(ct) &&
+	if (!nf_ct_is_confirmed(ct))
-	       !nf_ct_is_dying(ct);
+		return false;
+
+	/* load ct->timeout after is_confirmed() test.
+	 * Pairs with __nf_conntrack_confirm() which:
+	 * 1. Increases ct->timeout value
+	 * 2. Inserts ct into rcu hlist
+	 * 3. Sets the confirmed bit
+	 * 4. Unlocks the hlist lock
+	 */
+	smp_acquire__after_ctrl_dep();
+
+	return nf_ct_is_expired(ct) && !nf_ct_is_dying(ct);
 }
 
 #define	NF_CT_DAY	(86400 * HZ)

net/ipv4/udp.c

@@ -1414,12 +1414,12 @@ static bool udp_skb_has_head_state(struct sk_buff *skb)
 }
 
 /* fully reclaim rmem/fwd memory allocated for skb */
-static void udp_rmem_release(struct sock *sk, int size, int partial,
+static void udp_rmem_release(struct sock *sk, unsigned int size,
-			     bool rx_queue_lock_held)
+			     int partial, bool rx_queue_lock_held)
 {
 	struct udp_sock *up = udp_sk(sk);
 	struct sk_buff_head *sk_queue;
-	int amt;
+	unsigned int amt;
 
 	if (likely(partial)) {
 		up->forward_deficit += size;
@@ -1439,10 +1439,8 @@ static void udp_rmem_release(struct sock *sk, int size, int partial,
 	if (!rx_queue_lock_held)
 		spin_lock(&sk_queue->lock);
 
-
+	amt = (size + sk->sk_forward_alloc - partial) & ~(PAGE_SIZE - 1);
-	sk_forward_alloc_add(sk, size);
+	sk_forward_alloc_add(sk, size - amt);
-	amt = (sk->sk_forward_alloc - partial) & ~(PAGE_SIZE - 1);
-	sk_forward_alloc_add(sk, -amt);
 
 	if (amt)
 		__sk_mem_reduce_allocated(sk, amt >> PAGE_SHIFT);
@@ -1514,17 +1512,25 @@ static int udp_rmem_schedule(struct sock *sk, int size)
 int __udp_enqueue_schedule_skb(struct sock *sk, struct sk_buff *skb)
 {
 	struct sk_buff_head *list = &sk->sk_receive_queue;
-	int rmem, err = -ENOMEM;
+	unsigned int rmem, rcvbuf;
 	spinlock_t *busy = NULL;
-	int size, rcvbuf;
+	int size, err = -ENOMEM;
 
-	/* Immediately drop when the receive queue is full.
-	 * Always allow at least one packet.
-	 */
 	rmem = atomic_read(&sk->sk_rmem_alloc);
 	rcvbuf = READ_ONCE(sk->sk_rcvbuf);
+	size = skb->truesize;
+
+	/* Immediately drop when the receive queue is full.
+	 * Cast to unsigned int performs the boundary check for INT_MAX.
+	 */
+	if (rmem + size > rcvbuf) {
+		if (rcvbuf > INT_MAX >> 1)
+			goto drop;
+
+		/* Always allow at least one packet for small buffer. */
 		if (rmem > rcvbuf)
 			goto drop;
+	}
 
 	/* Under mem pressure, it might be helpful to help udp_recvmsg()
 	 * having linear skbs :
@@ -1534,10 +1540,10 @@ int __udp_enqueue_schedule_skb(struct sock *sk, struct sk_buff *skb)
 	 */
 	if (rmem > (rcvbuf >> 1)) {
 		skb_condense(skb);
-
+		size = skb->truesize;
 		busy = busylock_acquire(sk);
 	}
-	size = skb->truesize;
+
 	udp_set_dev_scratch(skb);
 
 	atomic_add(size, &sk->sk_rmem_alloc);
@@ -1624,7 +1630,7 @@ EXPORT_SYMBOL_GPL(skb_consume_udp);
 
 static struct sk_buff *__first_packet_length(struct sock *sk,
 					     struct sk_buff_head *rcvq,
-					     int *total)
+					     unsigned int *total)
 {
 	struct sk_buff *skb;
 
@@ -1657,8 +1663,8 @@ static int first_packet_length(struct sock *sk)
 {
 	struct sk_buff_head *rcvq = &udp_sk(sk)->reader_queue;
 	struct sk_buff_head *sk_queue = &sk->sk_receive_queue;
+	unsigned int total = 0;
 	struct sk_buff *skb;
-	int total = 0;
 	int res;
 
 	spin_lock_bh(&rcvq->lock);

net/netfilter/nf_conntrack_core.c

@@ -1121,6 +1121,12 @@ static int nf_ct_resolve_clash_harder(struct sk_buff *skb, u32 repl_idx)
 
 	hlist_nulls_add_head_rcu(&loser_ct->tuplehash[IP_CT_DIR_REPLY].hnnode,
 				 &nf_conntrack_hash[repl_idx]);
+	/* confirmed bit must be set after hlist add, not before:
+	 * loser_ct can still be visible to other cpu due to
+	 * SLAB_TYPESAFE_BY_RCU.
+	 */
+	smp_mb__before_atomic();
+	set_bit(IPS_CONFIRMED_BIT, &loser_ct->status);
 
 	NF_CT_STAT_INC(net, clash_resolve);
 	return NF_ACCEPT;
@@ -1257,8 +1263,6 @@ __nf_conntrack_confirm(struct sk_buff *skb)
 	 * user context, else we insert an already 'dead' hash, blocking
 	 * further use of that particular connection -JM.
 	 */
-	ct->status |= IPS_CONFIRMED;
-
 	if (unlikely(nf_ct_is_dying(ct))) {
 		NF_CT_STAT_INC(net, insert_failed);
 		goto dying;
@@ -1290,7 +1294,7 @@ chaintoolong:
 		}
 	}
 
-	/* Timer relative to confirmation time, not original
+	/* Timeout is relative to confirmation time, not original
 	   setting time, otherwise we'd get timer wrap in
 	   weird delay cases. */
 	ct->timeout += nfct_time_stamp;
@@ -1298,11 +1302,21 @@ chaintoolong:
 	__nf_conntrack_insert_prepare(ct);
 
 	/* Since the lookup is lockless, hash insertion must be done after
-	 * starting the timer and setting the CONFIRMED bit. The RCU barriers
+	 * setting ct->timeout. The RCU barriers guarantee that no other CPU
-	 * guarantee that no other CPU can find the conntrack before the above
+	 * can find the conntrack before the above stores are visible.
-	 * stores are visible.
 	 */
 	__nf_conntrack_hash_insert(ct, hash, reply_hash);
+
+	/* IPS_CONFIRMED unset means 'ct not (yet) in hash', conntrack lookups
+	 * skip entries that lack this bit.  This happens when a CPU is looking
+	 * at a stale entry that is being recycled due to SLAB_TYPESAFE_BY_RCU
+	 * or when another CPU encounters this entry right after the insertion
+	 * but before the set-confirm-bit below.  This bit must not be set until
+	 * after __nf_conntrack_hash_insert().
+	 */
+	smp_mb__before_atomic();
+	set_bit(IPS_CONFIRMED_BIT, &ct->status);
+
 	nf_conntrack_double_unlock(hash, reply_hash);
 	local_bh_enable();
 

net/sched/sch_hfsc.c

@@ -1648,10 +1648,16 @@ hfsc_dequeue(struct Qdisc *sch)
 		if (cl->qdisc->q.qlen != 0) {
 			/* update ed */
 			next_len = qdisc_peek_len(cl->qdisc);
+			/* Check queue length again since some qdisc implementations
+			 * (e.g., netem/codel) might empty the queue during the peek
+			 * operation.
+			 */
+			if (cl->qdisc->q.qlen != 0) {
 				if (realtime)
 					update_ed(cl, next_len);
 				else
 					update_d(cl, next_len);
+			}
 		} else {
 			/* the class becomes passive */
 			eltree_remove(cl);

net/tipc/topsrv.c

@@ -704,8 +704,10 @@ static void tipc_topsrv_stop(struct net *net)
 	for (id = 0; srv->idr_in_use; id++) {
 		con = idr_find(&srv->conn_idr, id);
 		if (con) {
+			conn_get(con);
 			spin_unlock_bh(&srv->idr_lock);
 			tipc_conn_close(con);
+			conn_put(con);
 			spin_lock_bh(&srv->idr_lock);
 		}
 	}

net/vmw_vsock/af_vsock.c

@@ -407,6 +407,8 @@ EXPORT_SYMBOL_GPL(vsock_enqueue_accept);
 
 static bool vsock_use_local_transport(unsigned int remote_cid)
 {
+	lockdep_assert_held(&vsock_register_mutex);
+
 	if (!transport_local)
 		return false;
 
@@ -464,6 +466,8 @@ int vsock_assign_transport(struct vsock_sock *vsk, struct vsock_sock *psk)
 
 	remote_flags = vsk->remote_addr.svm_flags;
 
+	mutex_lock(&vsock_register_mutex);
+
 	switch (sk->sk_type) {
 	case SOCK_DGRAM:
 		new_transport = transport_dgram;
@@ -479,12 +483,15 @@ int vsock_assign_transport(struct vsock_sock *vsk, struct vsock_sock *psk)
 			new_transport = transport_h2g;
 		break;
 	default:
-		return -ESOCKTNOSUPPORT;
+		ret = -ESOCKTNOSUPPORT;
+		goto err;
 	}
 
 	if (vsk->transport) {
-		if (vsk->transport == new_transport)
+		if (vsk->transport == new_transport) {
-			return 0;
+			ret = 0;
+			goto err;
+		}
 
 		/* transport->release() must be called with sock lock acquired.
 		 * This path can only be taken during vsock_connect(), where we
@@ -499,8 +506,16 @@ int vsock_assign_transport(struct vsock_sock *vsk, struct vsock_sock *psk)
 	/* We increase the module refcnt to prevent the transport unloading
 	 * while there are open sockets assigned to it.
 	 */
-	if (!new_transport || !try_module_get(new_transport->module))
+	if (!new_transport || !try_module_get(new_transport->module)) {
-		return -ENODEV;
+		ret = -ENODEV;
+		goto err;
+	}
+
+	/* It's safe to release the mutex after a successful try_module_get().
+	 * Whichever transport `new_transport` points at, it won't go away until
+	 * the last module_put() below or in vsock_deassign_transport().
+	 */
+	mutex_unlock(&vsock_register_mutex);
 
 	if (sk->sk_type == SOCK_SEQPACKET) {
 		if (!new_transport->seqpacket_allow ||
@@ -519,6 +534,9 @@ int vsock_assign_transport(struct vsock_sock *vsk, struct vsock_sock *psk)
 	vsk->transport = new_transport;
 
 	return 0;
+err:
+	mutex_unlock(&vsock_register_mutex);
+	return ret;
 }
 EXPORT_SYMBOL_GPL(vsock_assign_transport);