4.4.78-stable review

@@ -1004,6 +1004,8 @@ qdisc_create(struct net_device *dev, str
 
 		return sch;
 	}
+	/* ops->init() failed, we call ->destroy() like qdisc_create_dflt() */
+	ops->destroy(sch);
 err_out3:
 	dev_put(dev);
 	kfree((char *) sch - sch->padded);

@@ -636,7 +636,9 @@ static int hhf_init(struct Qdisc *sch, s
 			q->hhf_arrays[i] = hhf_zalloc(HHF_ARRAYS_LEN *
 						      sizeof(u32));
 			if (!q->hhf_arrays[i]) {
-				hhf_destroy(sch);
+				/* Note: hhf_destroy() will be called
+				 * by our caller.
+				 */
 				return -ENOMEM;
 			}
 		}

@@ -647,7 +649,9 @@ static int hhf_init(struct Qdisc *sch, s
 			q->hhf_valid_bits[i] = hhf_zalloc(HHF_ARRAYS_LEN /
 							  BITS_PER_BYTE);
 			if (!q->hhf_valid_bits[i]) {
-				hhf_destroy(sch);
+				/* Note: hhf_destroy() will be called
+				 * by our caller.
+				 */
 				return -ENOMEM;
 			}
 		}

@@ -52,7 +52,7 @@ static int mq_init(struct Qdisc *sch, st
 	/* pre-allocate qdiscs, attachment can't fail */
 	priv->qdiscs = kcalloc(dev->num_tx_queues, sizeof(priv->qdiscs[0]),
 			       GFP_KERNEL);
-	if (priv->qdiscs == NULL)
+	if (!priv->qdiscs)
 		return -ENOMEM;
 
 	for (ntx = 0; ntx < dev->num_tx_queues; ntx++) {

@@ -60,18 +60,14 @@ static int mq_init(struct Qdisc *sch, st
 		qdisc = qdisc_create_dflt(dev_queue, default_qdisc_ops,
 					  TC_H_MAKE(TC_H_MAJ(sch->handle),
 						    TC_H_MIN(ntx + 1)));
-		if (qdisc == NULL)
-			goto err;
+		if (!qdisc)
+			return -ENOMEM;
 		priv->qdiscs[ntx] = qdisc;
 		qdisc->flags |= TCQ_F_ONETXQUEUE | TCQ_F_NOPARENT;
 	}
 
 	sch->flags |= TCQ_F_MQROOT;
 	return 0;
-
-err:
-	mq_destroy(sch);
-	return -ENOMEM;
 }
 
 static void mq_attach(struct Qdisc *sch)

@@ -117,20 +117,17 @@ static int mqprio_init(struct Qdisc *sch
 	/* pre-allocate qdisc, attachment can't fail */
 	priv->qdiscs = kcalloc(dev->num_tx_queues, sizeof(priv->qdiscs[0]),
 			       GFP_KERNEL);
-	if (priv->qdiscs == NULL) {
-		err = -ENOMEM;
-		goto err;
-	}
+	if (!priv->qdiscs)
+		return -ENOMEM;
 
 	for (i = 0; i < dev->num_tx_queues; i++) {
 		dev_queue = netdev_get_tx_queue(dev, i);
 		qdisc = qdisc_create_dflt(dev_queue, default_qdisc_ops,
 					  TC_H_MAKE(TC_H_MAJ(sch->handle),
 						    TC_H_MIN(i + 1)));
-		if (qdisc == NULL) {
-			err = -ENOMEM;
-			goto err;
-		}
+		if (!qdisc)
+			return -ENOMEM;
+
 		priv->qdiscs[i] = qdisc;
 		qdisc->flags |= TCQ_F_ONETXQUEUE | TCQ_F_NOPARENT;
 	}

@@ -143,7 +140,7 @@ static int mqprio_init(struct Qdisc *sch
 		priv->hw_owned = 1;
 		err = dev->netdev_ops->ndo_setup_tc(dev, qopt->num_tc);
 		if (err)
-			goto err;
+			return err;
 	} else {
 		netdev_set_num_tc(dev, qopt->num_tc);
 		for (i = 0; i < qopt->num_tc; i++)

@@ -157,10 +154,6 @@ static int mqprio_init(struct Qdisc *sch
 
 	sch->flags |= TCQ_F_MQROOT;
 	return 0;
-
-err:
-	mqprio_destroy(sch);
-	return err;
 }
 
 static void mqprio_attach(struct Qdisc *sch)

@@ -742,9 +742,10 @@ static int sfq_init(struct Qdisc *sch, s
 	q->ht = sfq_alloc(sizeof(q->ht[0]) * q->divisor);
 	q->slots = sfq_alloc(sizeof(q->slots[0]) * q->maxflows);
 	if (!q->ht || !q->slots) {
-		sfq_destroy(sch);
+		/* Note: sfq_destroy() will be called by our caller */
 		return -ENOMEM;
 	}
+
 	for (i = 0; i < q->divisor; i++)
 		q->ht[i] = SFQ_EMPTY_SLOT;
 

@@ -1005,7 +1005,8 @@ qdisc_create(struct net_device *dev, str
 		return sch;
 	}
 	/* ops->init() failed, we call ->destroy() like qdisc_create_dflt() */
-	ops->destroy(sch);
+	if (ops->destroy)
+		ops->destroy(sch);
 err_out3:
 	dev_put(dev);
 	kfree((char *) sch - sch->padded);

@@ -539,6 +539,8 @@ static int ksz9031_read_status(struct ph
 	if ((regval & 0xFF) == 0xFF) {
 		phy_init_hw(phydev);
 		phydev->link = 0;
+		if (phydev->drv->config_intr && phy_interrupt_is_valid(phydev))
+			phydev->drv->config_intr(phydev);
 	}
 
 	return 0;

@@ -3143,6 +3143,7 @@ static int addrconf_notify(struct notifi
 {
 	struct net_device *dev = netdev_notifier_info_to_dev(ptr);
 	struct inet6_dev *idev = __in6_dev_get(dev);
+	struct net *net = dev_net(dev);
 	int run_pending = 0;
 	int err;
 

@@ -3158,7 +3159,7 @@ static int addrconf_notify(struct notifi
 	case NETDEV_CHANGEMTU:
 		/* if MTU under IPV6_MIN_MTU stop IPv6 on this interface. */
 		if (dev->mtu < IPV6_MIN_MTU) {
-			addrconf_ifdown(dev, 1);
+			addrconf_ifdown(dev, dev != net->loopback_dev);
 			break;
 		}
 

@@ -3271,7 +3272,7 @@ static int addrconf_notify(struct notifi
 			 * IPV6_MIN_MTU stop IPv6 on this interface.
 			 */
 			if (dev->mtu < IPV6_MIN_MTU)
-				addrconf_ifdown(dev, 1);
+				addrconf_ifdown(dev, dev != net->loopback_dev);
 		}
 		break;
 

@@ -907,7 +907,7 @@ static void decode_txts(struct dp83640_p
 	if (overflow) {
 		pr_debug("tx timestamp queue overflow, count %d\n", overflow);
 		while (skb) {
-			skb_complete_tx_timestamp(skb, NULL);
+			kfree_skb(skb);
 			skb = skb_dequeue(&dp83640->tx_queue);
 		}
 		return;

@@ -2263,6 +2263,8 @@ int tcp_disconnect(struct sock *sk, int
 	tcp_init_send_head(sk);
 	memset(&tp->rx_opt, 0, sizeof(tp->rx_opt));
 	__sk_dst_reset(sk);
+	dst_release(sk->sk_rx_dst);
+	sk->sk_rx_dst = NULL;
 	tcp_saved_syn_free(tp);
 
 	WARN_ON(inet->inet_num && !icsk->icsk_bind_hash);

@@ -7052,8 +7052,8 @@ struct rtnl_link_stats64 *dev_get_stats(
 	} else {
 		netdev_stats_to_stats64(storage, &dev->stats);
 	}
-	storage->rx_dropped += atomic_long_read(&dev->rx_dropped);
-	storage->tx_dropped += atomic_long_read(&dev->tx_dropped);
+	storage->rx_dropped += (unsigned long)atomic_long_read(&dev->rx_dropped);
+	storage->tx_dropped += (unsigned long)atomic_long_read(&dev->tx_dropped);
 	return storage;
 }
 EXPORT_SYMBOL(dev_get_stats);

@@ -754,6 +754,11 @@ static int check_xadd(struct verifier_en
 	if (err)
 		return err;
 
+	if (is_pointer_value(env, insn->src_reg)) {
+		verbose("R%d leaks addr into mem\n", insn->src_reg);
+		return -EACCES;
+	}
+
 	/* check whether atomic_add can read the memory */
 	err = check_mem_access(env, insn->dst_reg, insn->off,
 			       BPF_SIZE(insn->code), BPF_READ, -1);

@@ -1772,17 +1772,7 @@ struct inet6_ifaddr *ipv6_get_ifaddr(str
 
 static void addrconf_dad_stop(struct inet6_ifaddr *ifp, int dad_failed)
 {
-	if (ifp->flags&IFA_F_PERMANENT) {
-		spin_lock_bh(&ifp->lock);
-		addrconf_del_dad_work(ifp);
-		ifp->flags |= IFA_F_TENTATIVE;
-		if (dad_failed)
-			ifp->flags |= IFA_F_DADFAILED;
-		spin_unlock_bh(&ifp->lock);
-		if (dad_failed)
-			ipv6_ifa_notify(0, ifp);
-		in6_ifa_put(ifp);
-	} else if (ifp->flags&IFA_F_TEMPORARY) {
+	if (ifp->flags&IFA_F_TEMPORARY) {
 		struct inet6_ifaddr *ifpub;
 		spin_lock_bh(&ifp->lock);
 		ifpub = ifp->ifpub;

@@ -1795,6 +1785,16 @@ static void addrconf_dad_stop(struct ine
 			spin_unlock_bh(&ifp->lock);
 		}
 		ipv6_del_addr(ifp);
+	} else if (ifp->flags&IFA_F_PERMANENT || !dad_failed) {
+		spin_lock_bh(&ifp->lock);
+		addrconf_del_dad_work(ifp);
+		ifp->flags |= IFA_F_TENTATIVE;
+		if (dad_failed)
+			ifp->flags |= IFA_F_DADFAILED;
+		spin_unlock_bh(&ifp->lock);
+		if (dad_failed)
+			ipv6_ifa_notify(0, ifp);
+		in6_ifa_put(ifp);
 	} else {
 		ipv6_del_addr(ifp);
 	}

@@ -733,15 +733,15 @@ static int vrf_del_slave(struct net_devi
 static void vrf_dev_uninit(struct net_device *dev)
 {
 	struct net_vrf *vrf = netdev_priv(dev);
-	struct slave_queue *queue = &vrf->queue;
-	struct list_head *head = &queue->all_slaves;
-	struct slave *slave, *next;
+//	struct slave_queue *queue = &vrf->queue;
+//	struct list_head *head = &queue->all_slaves;
+//	struct slave *slave, *next;
 
 	vrf_rtable_destroy(vrf);
 	vrf_rt6_destroy(vrf);
 
-	list_for_each_entry_safe(slave, next, head, list)
-		vrf_del_slave(dev, slave->dev);
+//	list_for_each_entry_safe(slave, next, head, list)
+//		vrf_del_slave(dev, slave->dev);
 
 	free_percpu(dev->dstats);
 	dev->dstats = NULL;

@@ -914,6 +914,14 @@ static int vrf_validate(struct nlattr *t
 
 static void vrf_dellink(struct net_device *dev, struct list_head *head)
 {
+	struct net_vrf *vrf = netdev_priv(dev);
+	struct slave_queue *queue = &vrf->queue;
+	struct list_head *all_slaves = &queue->all_slaves;
+	struct slave *slave, *next;
+
+	list_for_each_entry_safe(slave, next, all_slaves, list)
+		vrf_del_slave(dev, slave->dev);
+
 	unregister_netdevice_queue(dev, head);
 }
 

@@ -78,7 +78,7 @@ int rds_tcp_accept_one(struct socket *so
 	struct inet_sock *inet;
 	struct rds_tcp_connection *rs_tcp;
 
-	ret = sock_create_kern(sock_net(sock->sk), sock->sk->sk_family,
+	ret = sock_create_lite(sock->sk->sk_family,
 			       sock->sk->sk_type, sock->sk->sk_protocol,
 			       &new_sock);
 	if (ret)

@@ -359,6 +359,7 @@ static const struct nla_policy nl80211_p
 	[NL80211_ATTR_SCAN_FLAGS] = { .type = NLA_U32 },
 	[NL80211_ATTR_P2P_CTWINDOW] = { .type = NLA_U8 },
 	[NL80211_ATTR_P2P_OPPPS] = { .type = NLA_U8 },
+	[NL80211_ATTR_LOCAL_MESH_POWER_MODE] = {. type = NLA_U32 },
 	[NL80211_ATTR_ACL_POLICY] = {. type = NLA_U32 },
 	[NL80211_ATTR_MAC_ADDRS] = { .type = NLA_NESTED },
 	[NL80211_ATTR_STA_CAPABILITY] = { .type = NLA_U16 },

@@ -5706,6 +5706,10 @@ static int validate_scan_freqs(struct nl
 	struct nlattr *attr1, *attr2;
 	int n_channels = 0, tmp1, tmp2;
 
+	nla_for_each_nested(attr1, freqs, tmp1)
+		if (nla_len(attr1) != sizeof(u32))
+			return 0;
+
 	nla_for_each_nested(attr1, freqs, tmp1) {
 		n_channels++;
 		/*

@@ -302,8 +302,7 @@ static const struct nla_policy nl80211_p
 	[NL80211_ATTR_WPA_VERSIONS] = { .type = NLA_U32 },
 	[NL80211_ATTR_PID] = { .type = NLA_U32 },
 	[NL80211_ATTR_4ADDR] = { .type = NLA_U8 },
-	[NL80211_ATTR_PMKID] = { .type = NLA_BINARY,
-				 .len = WLAN_PMKID_LEN },
+	[NL80211_ATTR_PMKID] = { .len = WLAN_PMKID_LEN },
 	[NL80211_ATTR_DURATION] = { .type = NLA_U32 },
 	[NL80211_ATTR_COOKIE] = { .type = NLA_U64 },
 	[NL80211_ATTR_TX_RATES] = { .type = NLA_NESTED },

@@ -632,6 +632,9 @@ static int gic_set_affinity(struct irq_d
 	int enabled;
 	u64 val;
 
+	if (cpu >= nr_cpu_ids)
+		return -EINVAL;
+
 	if (gic_irq_in_rdist(d))
 		return -EINVAL;
 

@@ -303,7 +303,7 @@ bad_area:
 		case 15:	/* Data TLB miss fault/Data page fault */
 			/* send SIGSEGV when outside of vma */
 			if (!vma ||
-			    address < vma->vm_start || address > vma->vm_end) {
+			    address < vma->vm_start || address >= vma->vm_end) {
 				si.si_signo = SIGSEGV;
 				si.si_code = SEGV_MAPERR;
 				break;

@@ -39,6 +39,8 @@ struct hppa_dma_ops {
 ** flush/purge and allocate "regular" cacheable pages for everything.
 */
 
+#define DMA_ERROR_CODE	(~(dma_addr_t)0)
+
 #ifdef CONFIG_PA11
 extern struct hppa_dma_ops pcxl_dma_ops;
 extern struct hppa_dma_ops pcx_dma_ops;

@@ -209,12 +211,13 @@ parisc_walk_tree(struct device *dev)
 			break;
 		}
 	}
-	BUG_ON(!dev->platform_data);
 	return dev->platform_data;
 }
-		
-#define GET_IOC(dev) (HBA_DATA(parisc_walk_tree(dev))->iommu)
-	
+
+#define GET_IOC(dev) ({					\
+	void *__pdata = parisc_walk_tree(dev);		\
+	__pdata ? HBA_DATA(__pdata)->iommu : NULL;	\
+})
 
 #ifdef CONFIG_IOMMU_CCIO
 struct parisc_device;

@@ -741,6 +741,8 @@ ccio_map_single(struct device *dev, void
 
 	BUG_ON(!dev);
 	ioc = GET_IOC(dev);
+	if (!ioc)
+		return DMA_ERROR_CODE;
 
 	BUG_ON(size <= 0);
 

@@ -805,6 +807,10 @@ ccio_unmap_single(struct device *dev, dm
 	
 	BUG_ON(!dev);
 	ioc = GET_IOC(dev);
+	if (!ioc) {
+		WARN_ON(!ioc);
+		return;
+	}
 
 	DBG_RUN("%s() iovp 0x%lx/%x\n",
 		__func__, (long)iova, size);

@@ -908,6 +914,8 @@ ccio_map_sg(struct device *dev, struct s
 	
 	BUG_ON(!dev);
 	ioc = GET_IOC(dev);
+	if (!ioc)
+		return 0;
 	
 	DBG_RUN_SG("%s() START %d entries\n", __func__, nents);
 

@@ -980,6 +988,10 @@ ccio_unmap_sg(struct device *dev, struct
 
 	BUG_ON(!dev);
 	ioc = GET_IOC(dev);
+	if (!ioc) {
+		WARN_ON(!ioc);
+		return;
+	}
 
 	DBG_RUN_SG("%s() START %d entries, %p,%x\n",
 		__func__, nents, sg_virt(sglist), sglist->length);

@@ -154,7 +154,10 @@ struct dino_device
 };
 
 /* Looks nice and keeps the compiler happy */
-#define DINO_DEV(d) ((struct dino_device *) d)
+#define DINO_DEV(d) ({				\
+	void *__pdata = d;			\
+	BUG_ON(!__pdata);			\
+	(struct dino_device *)__pdata; })
 
 
 /*

@@ -111,8 +111,10 @@ static u32 lba_t32;
 
 
 /* Looks nice and keeps the compiler happy */
-#define LBA_DEV(d) ((struct lba_device *) (d))
-
+#define LBA_DEV(d) ({				\
+	void *__pdata = d;			\
+	BUG_ON(!__pdata);			\
+	(struct lba_device *)__pdata; })
 
 /*
 ** Only allow 8 subsidiary busses per LBA

@@ -691,6 +691,8 @@ static int sba_dma_supported( struct dev
 		return 0;
 
 	ioc = GET_IOC(dev);
+	if (!ioc)
+		return 0;
 
 	/*
 	 * check if mask is >= than the current max IO Virt Address

@@ -722,6 +724,8 @@ sba_map_single(struct device *dev, void
 	int pide;
 
 	ioc = GET_IOC(dev);
+	if (!ioc)
+		return DMA_ERROR_CODE;
 
 	/* save offset bits */
 	offset = ((dma_addr_t) (long) addr) & ~IOVP_MASK;

@@ -803,6 +807,10 @@ sba_unmap_single(struct device *dev, dma
 	DBG_RUN("%s() iovp 0x%lx/%x\n", __func__, (long) iova, size);
 
 	ioc = GET_IOC(dev);
+	if (!ioc) {
+		WARN_ON(!ioc);
+		return;
+	}
 	offset = iova & ~IOVP_MASK;
 	iova ^= offset;        /* clear offset bits */
 	size += offset;

@@ -942,6 +950,8 @@ sba_map_sg(struct device *dev, struct sc
 	DBG_RUN_SG("%s() START %d entries\n", __func__, nents);
 
 	ioc = GET_IOC(dev);
+	if (!ioc)
+		return 0;
 
 	/* Fast path single entry scatterlists. */
 	if (nents == 1) {

@@ -1027,6 +1037,10 @@ sba_unmap_sg(struct device *dev, struct
 		__func__, nents, sg_virt(sglist), sglist->length);
 
 	ioc = GET_IOC(dev);
+	if (!ioc) {
+		WARN_ON(!ioc);
+		return;
+	}
 
 #ifdef SBA_COLLECT_STATS
 	ioc->usg_calls++;

@@ -49,15 +49,26 @@ static inline void load_context(mm_conte
 	mtctl(__space_to_prot(context), 8);
 }
 
-static inline void switch_mm(struct mm_struct *prev, struct mm_struct *next, struct task_struct *tsk)
+static inline void switch_mm_irqs_off(struct mm_struct *prev,
+		struct mm_struct *next, struct task_struct *tsk)
 {
-
 	if (prev != next) {
 		mtctl(__pa(next->pgd), 25);
 		load_context(next->context);
 	}
 }
 
+static inline void switch_mm(struct mm_struct *prev,
+		struct mm_struct *next, struct task_struct *tsk)
+{
+	unsigned long flags;
+
+	local_irq_save(flags);
+	switch_mm_irqs_off(prev, next, tsk);
+	local_irq_restore(flags);
+}
+#define switch_mm_irqs_off switch_mm_irqs_off
+
 #define deactivate_mm(tsk,mm)	do { } while (0)
 
 static inline void activate_mm(struct mm_struct *prev, struct mm_struct *next)

@@ -8,7 +8,7 @@
 #include <linux/utsname.h>
 #include <linux/compiler.h>
 
-#define MAX_LOCK_DEPTH 2000UL
+#define MAX_LOCK_DEPTH 255UL
 
 #define asmlinkage
 #define __visible

@@ -66,7 +66,7 @@ static inline int init_kernel_text(unsig
 	return 0;
 }
 
-int core_kernel_text(unsigned long addr)
+int notrace core_kernel_text(unsigned long addr)
 {
 	if (addr >= (unsigned long)_stext &&
 	    addr < (unsigned long)_etext)

@@ -44,6 +44,7 @@ struct list_lru_node {
 	/* for cgroup aware lrus points to per cgroup lists, otherwise NULL */
 	struct list_lru_memcg	*memcg_lrus;
 #endif
+	long nr_items;
 } ____cacheline_aligned_in_smp;
 
 struct list_lru {

@@ -117,6 +117,7 @@ bool list_lru_add(struct list_lru *lru,
 		l = list_lru_from_kmem(nlru, item);
 		list_add_tail(item, &l->list);
 		l->nr_items++;
+		nlru->nr_items++;
 		spin_unlock(&nlru->lock);
 		return true;
 	}

@@ -136,6 +137,7 @@ bool list_lru_del(struct list_lru *lru,
 		l = list_lru_from_kmem(nlru, item);
 		list_del_init(item);
 		l->nr_items--;
+		nlru->nr_items--;
 		spin_unlock(&nlru->lock);
 		return true;
 	}

@@ -183,15 +185,10 @@ EXPORT_SYMBOL_GPL(list_lru_count_one);
 
 unsigned long list_lru_count_node(struct list_lru *lru, int nid)
 {
-	long count = 0;
-	int memcg_idx;
+	struct list_lru_node *nlru;
 
-	count += __list_lru_count_one(lru, nid, -1);
-	if (list_lru_memcg_aware(lru)) {
-		for_each_memcg_cache_index(memcg_idx)
-			count += __list_lru_count_one(lru, nid, memcg_idx);
-	}
-	return count;
+	nlru = &lru->node[nid];
+	return nlru->nr_items;
 }
 EXPORT_SYMBOL_GPL(list_lru_count_node);
 

@@ -226,6 +223,7 @@ restart:
 			assert_spin_locked(&nlru->lock);
 		case LRU_REMOVED:
 			isolated++;
+			nlru->nr_items--;
 			/*
 			 * If the lru lock has been dropped, our list
 			 * traversal is now invalid and so we have to

@@ -1128,11 +1128,12 @@ void shrink_dcache_sb(struct super_block
 		LIST_HEAD(dispose);
 
 		freed = list_lru_walk(&sb->s_dentry_lru,
-			dentry_lru_isolate_shrink, &dispose, UINT_MAX);
+			dentry_lru_isolate_shrink, &dispose, 1024);
 
 		this_cpu_sub(nr_dentry_unused, freed);
 		shrink_dentry_list(&dispose);
-	} while (freed > 0);
+		cond_resched();
+	} while (list_lru_count(&sb->s_dentry_lru) > 0);
 }
 EXPORT_SYMBOL(shrink_dcache_sb);
 

@@ -245,12 +245,13 @@ extern int force_personality32;
 #define CORE_DUMP_USE_REGSET
 #define ELF_EXEC_PAGESIZE	4096
 
-/* This is the location that an ET_DYN program is loaded if exec'ed.  Typical
-   use of this is to invoke "./ld.so someprog" to test out a new version of
-   the loader.  We need to make sure that it is out of the way of the program
-   that it will "exec", and that there is sufficient room for the brk.  */
-
-#define ELF_ET_DYN_BASE		(TASK_SIZE / 3 * 2)
+/*
+ * This is the base location for PIE (ET_DYN with INTERP) loads. On
+ * 64-bit, this is raised to 4GB to leave the entire 32-bit address
+ * space open for things that want to use the area for 32-bit pointers.
+ */
+#define ELF_ET_DYN_BASE		(mmap_is_ia32() ? 0x000400000UL : \
+						  0x100000000UL)
 
 /* This yields a mask that user programs can use to figure out what
    instruction set this CPU supports.  This could be done in user space,

68 lines not on this page · raw message

@@ -112,12 +112,8 @@ int dump_task_regs(struct task_struct *t
 #define CORE_DUMP_USE_REGSET
 #define ELF_EXEC_PAGESIZE	4096
 
-/* This is the location that an ET_DYN program is loaded if exec'ed.  Typical
-   use of this is to invoke "./ld.so someprog" to test out a new version of
-   the loader.  We need to make sure that it is out of the way of the program
-   that it will "exec", and that there is sufficient room for the brk.  */
-
-#define ELF_ET_DYN_BASE	(TASK_SIZE / 3 * 2)
+/* This is the base location for PIE (ET_DYN with INTERP) loads. */
+#define ELF_ET_DYN_BASE		0x400000UL
 
 /* When the program starts, a1 contains a pointer to a function to be 
    registered with atexit, as per the SVR4 ABI.  A value of 0 means we 

@@ -120,12 +120,11 @@ typedef struct user_fpsimd_state elf_fpr
 #define ELF_EXEC_PAGESIZE	PAGE_SIZE
 
 /*
- * This is the location that an ET_DYN program is loaded if exec'ed.  Typical
- * use of this is to invoke "./ld.so someprog" to test out a new version of
- * the loader.  We need to make sure that it is out of the way of the program
- * that it will "exec", and that there is sufficient room for the brk.
+ * This is the base location for PIE (ET_DYN with INTERP) loads. On
+ * 64-bit, this is raised to 4GB to leave the entire 32-bit address
+ * space open for things that want to use the area for 32-bit pointers.
  */
-#define ELF_ET_DYN_BASE	(2 * TASK_SIZE_64 / 3)
+#define ELF_ET_DYN_BASE		0x100000000UL
 
 /*
  * When the program starts, a1 contains a pointer to a function to be

@@ -165,7 +164,8 @@ extern int arch_setup_additional_pages(s
 
 #ifdef CONFIG_COMPAT
 
-#define COMPAT_ELF_ET_DYN_BASE		(2 * TASK_SIZE_32 / 3)
+/* PIE load location for compat arm. Must match ARM ELF_ET_DYN_BASE. */
+#define COMPAT_ELF_ET_DYN_BASE		0x000400000UL
 
 /* AArch32 registers. */
 #define COMPAT_ELF_NGREG		18

@@ -23,12 +23,13 @@
 #define CORE_DUMP_USE_REGSET
 #define ELF_EXEC_PAGESIZE	PAGE_SIZE
 
-/* This is the location that an ET_DYN program is loaded if exec'ed.  Typical
-   use of this is to invoke "./ld.so someprog" to test out a new version of
-   the loader.  We need to make sure that it is out of the way of the program
-   that it will "exec", and that there is sufficient room for the brk.  */
-
-#define ELF_ET_DYN_BASE	0x20000000
+/*
+ * This is the base location for PIE (ET_DYN with INTERP) loads. On
+ * 64-bit, this is raised to 4GB to leave the entire 32-bit address
+ * space open for things that want to use the area for 32-bit pointers.
+ */
+#define ELF_ET_DYN_BASE		(is_32bit_task() ? 0x000400000UL : \
+						   0x100000000UL)
 
 #define ELF_CORE_EFLAGS (is_elf2_task() ? 2 : 0)
 

@@ -154,14 +154,13 @@ extern unsigned int vdso_enabled;
 #define CORE_DUMP_USE_REGSET
 #define ELF_EXEC_PAGESIZE	4096
 
-/* This is the location that an ET_DYN program is loaded if exec'ed.  Typical
-   use of this is to invoke "./ld.so someprog" to test out a new version of
-   the loader.  We need to make sure that it is out of the way of the program
-   that it will "exec", and that there is sufficient room for the brk. 64-bit
-   tasks are aligned to 4GB. */
-#define ELF_ET_DYN_BASE (is_32bit_task() ? \
-				(STACK_TOP / 3 * 2) : \
-				(STACK_TOP / 3 * 2) & ~((1UL << 32) - 1))
+/*
+ * This is the base location for PIE (ET_DYN with INTERP) loads. On
+ * 64-bit, this is raised to 4GB to leave the entire 32-bit address
+ * space open for things that want to use the area for 32-bit pointers.
+ */
+#define ELF_ET_DYN_BASE		(is_compat_task() ? 0x000400000UL : \
+						    0x100000000UL)
 
 /* This yields a mask that user programs can use to figure out what
    instruction set this CPU supports. */

@@ -206,8 +206,7 @@ static struct page *get_arg_page(struct
 
 	if (write) {
 		unsigned long size = bprm->vma->vm_end - bprm->vma->vm_start;
-		unsigned long ptr_size;
-		struct rlimit *rlim;
+		unsigned long ptr_size, limit;
 
 		/*
 		 * Since the stack will hold pointers to the strings, we

@@ -236,14 +235,16 @@ static struct page *get_arg_page(struct
 			return page;
 
 		/*
-		 * Limit to 1/4-th the stack size for the argv+env strings.
+		 * Limit to 1/4 of the max stack size or 3/4 of _STK_LIM
+		 * (whichever is smaller) for the argv+env strings.
 		 * This ensures that:
 		 *  - the remaining binfmt code will not run out of stack space,
 		 *  - the program will have a reasonable amount of stack left
 		 *    to work from.
 		 */
-		rlim = current->signal->rlim;
-		if (size > READ_ONCE(rlim[RLIMIT_STACK].rlim_cur) / 4)
+		limit = _STK_LIM / 4 * 3;
+		limit = min(limit, rlimit(RLIMIT_STACK) / 4);
+		if (size > limit)
 			goto fail;
 	}
 

@@ -2708,13 +2708,13 @@ int tioclinux(struct tty_struct *tty, un
 	 * related to the kernel should not use this.
 	 */
 			data = vt_get_shift_state();
-			ret = __put_user(data, p);
+			ret = put_user(data, p);
 			break;
 		case TIOCL_GETMOUSEREPORTING:
 			console_lock();	/* May be overkill */
 			data = mouse_reporting();
 			console_unlock();
-			ret = __put_user(data, p);
+			ret = put_user(data, p);
 			break;
 		case TIOCL_SETVESABLANK:
 			console_lock();

@@ -2723,7 +2723,7 @@ int tioclinux(struct tty_struct *tty, un
 			break;
 		case TIOCL_GETKMSGREDIRECT:
 			data = vt_get_kmsg_redirect();
-			ret = __put_user(data, p);
+			ret = put_user(data, p);
 			break;
 		case TIOCL_SETKMSGREDIRECT:
 			if (!capable(CAP_SYS_ADMIN)) {

@@ -57,6 +57,7 @@ struct mount {
 	struct mnt_namespace *mnt_ns;	/* containing namespace */
 	struct mountpoint *mnt_mp;	/* where is it mounted */
 	struct hlist_node mnt_mp_list;	/* list mounts with the same mountpoint */
+	struct list_head mnt_reparent;	/* reparent list entry */
 #ifdef CONFIG_FSNOTIFY
 	struct hlist_head mnt_fsnotify_marks;
 	__u32 mnt_fsnotify_mask;

@@ -237,6 +237,7 @@ static struct mount *alloc_vfsmnt(const
 		INIT_LIST_HEAD(&mnt->mnt_slave_list);
 		INIT_LIST_HEAD(&mnt->mnt_slave);
 		INIT_HLIST_NODE(&mnt->mnt_mp_list);
+		INIT_LIST_HEAD(&mnt->mnt_reparent);
 #ifdef CONFIG_FSNOTIFY
 		INIT_HLIST_HEAD(&mnt->mnt_fsnotify_marks);
 #endif

@@ -441,7 +441,7 @@ static void mark_umount_candidates(struc
  * NOTE: unmounting 'mnt' naturally propagates to all other mounts its
  * parent propagates to.
  */
-static void __propagate_umount(struct mount *mnt)
+static void __propagate_umount(struct mount *mnt, struct list_head *to_reparent)
 {
 	struct mount *parent = mnt->mnt_parent;
 	struct mount *m;

@@ -466,17 +466,38 @@ static void __propagate_umount(struct mo
 		 */
 		topper = find_topper(child);
 		if (topper)
-			mnt_change_mountpoint(child->mnt_parent, child->mnt_mp,
-					      topper);
+			list_add_tail(&topper->mnt_reparent, to_reparent);
 
-		if (list_empty(&child->mnt_mounts)) {
+		if (topper || list_empty(&child->mnt_mounts)) {
 			list_del_init(&child->mnt_child);
+			list_del_init(&child->mnt_reparent);
 			child->mnt.mnt_flags |= MNT_UMOUNT;
 			list_move_tail(&child->mnt_list, &mnt->mnt_list);
 		}
 	}
 }
 
+static void reparent_mounts(struct list_head *to_reparent)
+{
+	while (!list_empty(to_reparent)) {
+		struct mount *mnt, *parent;
+		struct mountpoint *mp;
+
+		mnt = list_first_entry(to_reparent, struct mount, mnt_reparent);
+		list_del_init(&mnt->mnt_reparent);
+
+		/* Where should this mount be reparented to? */
+		mp = mnt->mnt_mp;
+		parent = mnt->mnt_parent;
+		while (parent->mnt.mnt_flags & MNT_UMOUNT) {
+			mp = parent->mnt_mp;
+			parent = parent->mnt_parent;
+		}
+
+		mnt_change_mountpoint(parent, mp, mnt);
+	}
+}
+
 /*
  * collect all mounts that receive propagation from the mount in @list,
  * and return these additional mounts in the same list.

@@ -487,11 +508,15 @@ static void __propagate_umount(struct mo
 int propagate_umount(struct list_head *list)
 {
 	struct mount *mnt;
+	LIST_HEAD(to_reparent);
 
 	list_for_each_entry_reverse(mnt, list, mnt_list)
 		mark_umount_candidates(mnt);
 
 	list_for_each_entry(mnt, list, mnt_list)
-		__propagate_umount(mnt);
+		__propagate_umount(mnt, &to_reparent);
+
+	reparent_mounts(&to_reparent);
+
 	return 0;
 }

@@ -57,7 +57,7 @@ struct mount {
 	struct mnt_namespace *mnt_ns;	/* containing namespace */
 	struct mountpoint *mnt_mp;	/* where is it mounted */
 	struct hlist_node mnt_mp_list;	/* list mounts with the same mountpoint */
-	struct list_head mnt_reparent;	/* reparent list entry */
+	struct list_head mnt_umounting; /* list entry for umount propagation */
 #ifdef CONFIG_FSNOTIFY
 	struct hlist_head mnt_fsnotify_marks;
 	__u32 mnt_fsnotify_mask;

@@ -237,7 +237,7 @@ static struct mount *alloc_vfsmnt(const
 		INIT_LIST_HEAD(&mnt->mnt_slave_list);
 		INIT_LIST_HEAD(&mnt->mnt_slave);
 		INIT_HLIST_NODE(&mnt->mnt_mp_list);
-		INIT_LIST_HEAD(&mnt->mnt_reparent);
+		INIT_LIST_HEAD(&mnt->mnt_umounting);
 #ifdef CONFIG_FSNOTIFY
 		INIT_HLIST_HEAD(&mnt->mnt_fsnotify_marks);
 #endif

190 lines not on this page · raw message

@@ -24,6 +24,11 @@ static inline struct mount *first_slave(
 	return list_entry(p->mnt_slave_list.next, struct mount, mnt_slave);
 }
 
+static inline struct mount *last_slave(struct mount *p)
+{
+	return list_entry(p->mnt_slave_list.prev, struct mount, mnt_slave);
+}
+
 static inline struct mount *next_slave(struct mount *p)
 {
 	return list_entry(p->mnt_slave.next, struct mount, mnt_slave);

@@ -164,6 +169,19 @@ static struct mount *propagation_next(st
 	}
 }
 
+static struct mount *skip_propagation_subtree(struct mount *m,
+						struct mount *origin)
+{
+	/*
+	 * Advance m such that propagation_next will not return
+	 * the slaves of m.
+	 */
+	if (!IS_MNT_NEW(m) && !list_empty(&m->mnt_slave_list))
+		m = last_slave(m);
+
+	return m;
+}
+
 static struct mount *next_group(struct mount *m, struct mount *origin)
 {
 	while (1) {

@@ -507,6 +525,15 @@ static void restore_mounts(struct list_h
 	}
 }
 
+static void cleanup_umount_visitations(struct list_head *visited)
+{
+	while (!list_empty(visited)) {
+		struct mount *mnt =
+			list_first_entry(visited, struct mount, mnt_umounting);
+		list_del_init(&mnt->mnt_umounting);
+	}
+}
+
 /*
  * collect all mounts that receive propagation from the mount in @list,
  * and return these additional mounts in the same list.

@@ -519,11 +546,23 @@ int propagate_umount(struct list_head *l
 	struct mount *mnt;
 	LIST_HEAD(to_restore);
 	LIST_HEAD(to_umount);
+	LIST_HEAD(visited);
 
-	list_for_each_entry(mnt, list, mnt_list) {
+	/* Find candidates for unmounting */
+	list_for_each_entry_reverse(mnt, list, mnt_list) {
 		struct mount *parent = mnt->mnt_parent;
 		struct mount *m;
 
+		/*
+		 * If this mount has already been visited it is known that it's
+		 * entire peer group and all of their slaves in the propagation
+		 * tree for the mountpoint has already been visited and there is
+		 * no need to visit them again.
+		 */
+		if (!list_empty(&mnt->mnt_umounting))
+			continue;
+
+		list_add_tail(&mnt->mnt_umounting, &visited);
 		for (m = propagation_next(parent, parent); m;
 		     m = propagation_next(m, parent)) {
 			struct mount *child = __lookup_mnt(&m->mnt,

@@ -531,6 +570,27 @@ int propagate_umount(struct list_head *l
 			if (!child)
 				continue;
 
+			if (!list_empty(&child->mnt_umounting)) {
+				/*
+				 * If the child has already been visited it is
+				 * know that it's entire peer group and all of
+				 * their slaves in the propgation tree for the
+				 * mountpoint has already been visited and there
+				 * is no need to visit this subtree again.
+				 */
+				m = skip_propagation_subtree(m, parent);
+				continue;
+			} else if (child->mnt.mnt_flags & MNT_UMOUNT) {
+				/*
+				 * We have come accross an partially unmounted
+				 * mount in list that has not been visited yet.
+				 * Remember it has been visited and continue
+				 * about our merry way.
+				 */
+				list_add_tail(&child->mnt_umounting, &visited);
+				continue;
+			}
+
 			/* Check the child and parents while progress is made */
 			while (__propagate_umount(child,
 						  &to_umount, &to_restore)) {

@@ -544,6 +604,7 @@ int propagate_umount(struct list_head *l
 
 	umount_list(&to_umount, &to_restore);
 	restore_mounts(&to_restore);
+	cleanup_umount_visitations(&visited);
 	list_splice_tail(&to_umount, list);
 
 	return 0;

@@ -138,9 +138,6 @@ static void chdir_to_tmpfs(void)
 
 	if (chdir(cwd) != 0)
 		err(1, "chdir to private tmpfs");
-
-	if (umount2(".", MNT_DETACH) != 0)
-		err(1, "detach private tmpfs");
 }
 
 static void copy_fromat_to(int fromfd, const char *fromname, const char *toname)

@@ -248,7 +245,7 @@ static int do_tests(int uid, const char
 			err(1, "chown");
 		if (chmod("validate_cap_sgidnonroot", S_ISGID | 0710) != 0)
 			err(1, "chmod");
-}
+	}
 
 	capng_get_caps_process();
 

@@ -384,7 +381,7 @@ static int do_tests(int uid, const char
 	} else {
 		printf("[RUN]\tNon-root +ia, sgidnonroot => i\n");
 		exec_other_validate_cap("./validate_cap_sgidnonroot",
-						false, false, true, false);
+					false, false, true, false);
 
 		if (fork_wait()) {
 			printf("[RUN]\tNon-root +ia, sgidroot => i\n");

@@ -49,7 +49,7 @@ struct tpm_chip *tpm_chip_find_get(int c
 		if (chip_num != TPM_ANY_NUM && chip_num != pos->dev_num)
 			continue;
 
-		if (try_module_get(pos->pdev->driver->owner)) {
+		if (try_module_get(pos->dev.parent->driver->owner)) {
 			chip = pos;
 			break;
 		}

@@ -112,13 +112,11 @@ struct tpm_chip *tpmm_chip_alloc(struct
 
 	scnprintf(chip->devname, sizeof(chip->devname), "tpm%d", chip->dev_num);
 
-	chip->pdev = dev;
-
 	dev_set_drvdata(dev, chip);
 
 	chip->dev.class = tpm_class;
 	chip->dev.release = tpm_dev_release;
-	chip->dev.parent = chip->pdev;
+	chip->dev.parent = dev;
 #ifdef CONFIG_ACPI
 	chip->dev.groups = chip->groups;
 #endif

@@ -133,7 +131,7 @@ struct tpm_chip *tpmm_chip_alloc(struct
 	device_initialize(&chip->dev);
 
 	cdev_init(&chip->cdev, &tpm_fops);
-	chip->cdev.owner = chip->pdev->driver->owner;
+	chip->cdev.owner = dev->driver->owner;
 	chip->cdev.kobj.parent = &chip->dev.kobj;
 
 	devm_add_action(dev, (void (*)(void *)) put_device, &chip->dev);

@@ -236,9 +234,8 @@ int tpm_chip_register(struct tpm_chip *c
 	chip->flags |= TPM_CHIP_FLAG_REGISTERED;
 
 	if (!(chip->flags & TPM_CHIP_FLAG_TPM2)) {
-		rc = __compat_only_sysfs_link_entry_to_kobj(&chip->pdev->kobj,
-							    &chip->dev.kobj,
-							    "ppi");
+		rc = __compat_only_sysfs_link_entry_to_kobj(
+		    &chip->dev.parent->kobj, &chip->dev.kobj, "ppi");
 		if (rc && rc != -ENOENT) {
 			tpm_chip_unregister(chip);
 			return rc;

@@ -273,7 +270,7 @@ void tpm_chip_unregister(struct tpm_chip
 	synchronize_rcu();
 
 	if (!(chip->flags & TPM_CHIP_FLAG_TPM2))
-		sysfs_remove_link(&chip->pdev->kobj, "ppi");
+		sysfs_remove_link(&chip->dev.parent->kobj, "ppi");
 
 	tpm1_chip_unregister(chip);
 	tpm_del_char_device(chip);

@@ -61,7 +61,7 @@ static int tpm_open(struct inode *inode,
 	 * by the check of is_open variable, which is protected
 	 * by driver_lock. */
 	if (test_and_set_bit(0, &chip->is_open)) {
-		dev_dbg(chip->pdev, "Another process owns this TPM\n");
+		dev_dbg(&chip->dev, "Another process owns this TPM\n");
 		return -EBUSY;
 	}
 

@@ -79,7 +79,6 @@ static int tpm_open(struct inode *inode,
 	INIT_WORK(&priv->work, timeout_work);
 
 	file->private_data = priv;
-	get_device(chip->pdev);
 	return 0;
 }
 

@@ -166,7 +165,6 @@ static int tpm_release(struct inode *ino
 	file->private_data = NULL;
 	atomic_set(&priv->data_pending, 0);
 	clear_bit(0, &priv->chip->is_open);
-	put_device(priv->chip->pdev);
 	kfree(priv);
 	return 0;
 }

@@ -343,7 +343,7 @@ ssize_t tpm_transmit(struct tpm_chip *ch
 	if (count == 0)
 		return -ENODATA;
 	if (count > bufsiz) {
-		dev_err(chip->pdev,
+		dev_err(&chip->dev,
 			"invalid count value %x %zx\n", count, bufsiz);
 		return -E2BIG;
 	}

@@ -353,7 +353,7 @@ ssize_t tpm_transmit(struct tpm_chip *ch
 
 	rc = chip->ops->send(chip, (u8 *) buf, count);
 	if (rc < 0) {
-		dev_err(chip->pdev,
+		dev_err(&chip->dev,
 			"tpm_transmit: tpm_send: error %zd\n", rc);
 		goto out;
 	}

@@ -372,7 +372,7 @@ ssize_t tpm_transmit(struct tpm_chip *ch
 			goto out_recv;
 
 		if (chip->ops->req_canceled(chip, status)) {
-			dev_err(chip->pdev, "Operation Canceled\n");
+			dev_err(&chip->dev, "Operation Canceled\n");
 			rc = -ECANCELED;
 			goto out;
 		}

@@ -382,14 +382,14 @@ ssize_t tpm_transmit(struct tpm_chip *ch
 	} while (time_before(jiffies, stop));
 
 	chip->ops->cancel(chip);
-	dev_err(chip->pdev, "Operation Timed out\n");
+	dev_err(&chip->dev, "Operation Timed out\n");
 	rc = -ETIME;
 	goto out;
 
 out_recv:
 	rc = chip->ops->recv(chip, (u8 *) buf, bufsiz);
 	if (rc < 0)
-		dev_err(chip->pdev,
+		dev_err(&chip->dev,
 			"tpm_transmit: tpm_recv: error %zd\n", rc);
 out:
 	if (!(flags & TPM_TRANSMIT_UNLOCKED))

@@ -416,7 +416,7 @@ ssize_t tpm_transmit_cmd(struct tpm_chip
 
 	err = be32_to_cpu(header->return_code);
 	if (err != 0 && desc)
-		dev_err(chip->pdev, "A TPM error (%d) occurred %s\n", err,
+		dev_err(&chip->dev, "A TPM error (%d) occurred %s\n", err,
 			desc);
 
 	return err;

@@ -514,7 +514,7 @@ int tpm_get_timeouts(struct tpm_chip *ch
 	if (rc == TPM_ERR_INVALID_POSTINIT) {
 		/* The TPM is not started, we are the first to talk to it.
 		   Execute a startup command. */
-		dev_info(chip->pdev, "Issuing TPM_STARTUP");
+		dev_info(&chip->dev, "Issuing TPM_STARTUP");
 		if (tpm_startup(chip, TPM_ST_CLEAR))
 			return rc;
 

@@ -526,7 +526,7 @@ int tpm_get_timeouts(struct tpm_chip *ch
 				      0, NULL);
 	}
 	if (rc) {
-		dev_err(chip->pdev,
+		dev_err(&chip->dev,
 			"A TPM error (%zd) occurred attempting to determine the timeouts\n",
 			rc);
 		goto duration;

@@ -565,7 +565,7 @@ int tpm_get_timeouts(struct tpm_chip *ch
 
 	/* Report adjusted timeouts */
 	if (chip->vendor.timeout_adjusted) {
-		dev_info(chip->pdev,
+		dev_info(&chip->dev,
 			 HW_ERR "Adjusting reported timeouts: A %lu->%luus B %lu->%luus C %lu->%luus D %lu->%luus\n",
 			 old_timeout[0], new_timeout[0],
 			 old_timeout[1], new_timeout[1],

@@ -612,7 +612,7 @@ duration:
 		chip->vendor.duration[TPM_MEDIUM] *= 1000;
 		chip->vendor.duration[TPM_LONG] *= 1000;
 		chip->vendor.duration_adjusted = true;
-		dev_info(chip->pdev, "Adjusting TPM timeout parameters.");
+		dev_info(&chip->dev, "Adjusting TPM timeout parameters.");
 	}
 	return 0;
 }

@@ -802,7 +802,9 @@ int tpm_do_selftest(struct tpm_chip *chi
 		 * around 300ms while the self test is ongoing, keep trying
 		 * until the self test duration expires. */
 		if (rc == -ETIME) {
-			dev_info(chip->pdev, HW_ERR "TPM command timed out during continue self test");
+			dev_info(
+			    &chip->dev, HW_ERR
+			    "TPM command timed out during continue self test");
 			msleep(delay_msec);
 			continue;
 		}

@@ -812,7 +814,7 @@ int tpm_do_selftest(struct tpm_chip *chi
 
 		rc = be32_to_cpu(cmd.header.out.return_code);
 		if (rc == TPM_ERR_DISABLED || rc == TPM_ERR_DEACTIVATED) {
-			dev_info(chip->pdev,
+			dev_info(&chip->dev,
 				 "TPM is disabled/deactivated (0x%X)\n", rc);
 			/* TPM is disabled and/or deactivated; driver can
 			 * proceed and TPM does handle commands for

@@ -966,10 +968,10 @@ int tpm_pm_suspend(struct device *dev)
 	}
 
 	if (rc)
-		dev_err(chip->pdev,
+		dev_err(&chip->dev,
 			"Error (%d) sending savestate before suspend\n", rc);
 	else if (try > 0)
-		dev_warn(chip->pdev, "TPM savestate took %dms\n",
+		dev_warn(&chip->dev, "TPM savestate took %dms\n",
 			 try * TPM_TIMEOUT_RETRY);
 
 	return rc;

@@ -284,16 +284,16 @@ static const struct attribute_group tpm_
 int tpm_sysfs_add_device(struct tpm_chip *chip)
 {
 	int err;
-	err = sysfs_create_group(&chip->pdev->kobj,
+	err = sysfs_create_group(&chip->dev.parent->kobj,
 				 &tpm_dev_group);
 
 	if (err)
-		dev_err(chip->pdev,
+		dev_err(&chip->dev,
 			"failed to create sysfs attributes, %d\n", err);
 	return err;
 }
 
 void tpm_sysfs_del_device(struct tpm_chip *chip)
 {
-	sysfs_remove_group(&chip->pdev->kobj, &tpm_dev_group);
+	sysfs_remove_group(&chip->dev.parent->kobj, &tpm_dev_group);
 }

@@ -171,7 +171,6 @@ enum tpm_chip_flags {
 };
 
 struct tpm_chip {
-	struct device *pdev;	/* Device stuff */
 	struct device dev;
 	struct cdev cdev;
 

@@ -203,7 +202,7 @@ struct tpm_chip {
 
 static inline void tpm_chip_put(struct tpm_chip *chip)
 {
-	module_put(chip->pdev->driver->owner);
+	module_put(chip->dev.parent->driver->owner);
 }
 
 static inline int tpm_read_index(int base, int index)

@@ -570,7 +570,7 @@ static void tpm2_flush_context_cmd(struc
 
 	rc = tpm_buf_init(&buf, TPM2_ST_NO_SESSIONS, TPM2_CC_FLUSH_CONTEXT);
 	if (rc) {
-		dev_warn(chip->pdev, "0x%08x was not flushed, out of memory\n",
+		dev_warn(&chip->dev, "0x%08x was not flushed, out of memory\n",
 			 handle);
 		return;
 	}

@@ -580,7 +580,7 @@ static void tpm2_flush_context_cmd(struc
 	rc = tpm_transmit_cmd(chip, buf.data, PAGE_SIZE, flags,
 			      "flushing context");
 	if (rc)
-		dev_warn(chip->pdev, "0x%08x was not flushed, rc=%d\n", handle,
+		dev_warn(&chip->dev, "0x%08x was not flushed, rc=%d\n", handle,
 			 rc);
 
 	tpm_buf_destroy(&buf);

@@ -753,7 +753,7 @@ void tpm2_shutdown(struct tpm_chip *chip
 	 * except print the error code on a system failure.
 	 */
 	if (rc < 0)
-		dev_warn(chip->pdev, "transmit returned %d while stopping the TPM",
+		dev_warn(&chip->dev, "transmit returned %d while stopping the TPM",
 			 rc);
 }
 EXPORT_SYMBOL_GPL(tpm2_shutdown);

@@ -820,7 +820,7 @@ static int tpm2_start_selftest(struct tp
 	 * immediately. This is a workaround for that.
 	 */
 	if (rc == TPM2_RC_TESTING) {
-		dev_warn(chip->pdev, "Got RC_TESTING, ignoring\n");
+		dev_warn(&chip->dev, "Got RC_TESTING, ignoring\n");
 		rc = 0;
 	}
 

@@ -49,7 +49,7 @@ static int tpm_atml_recv(struct tpm_chip
 	for (i = 0; i < 6; i++) {
 		status = ioread8(chip->vendor.iobase + 1);
 		if ((status & ATML_STATUS_DATA_AVAIL) == 0) {
-			dev_err(chip->pdev, "error reading header\n");
+			dev_err(&chip->dev, "error reading header\n");
 			return -EIO;
 		}
 		*buf++ = ioread8(chip->vendor.iobase);

@@ -60,12 +60,12 @@ static int tpm_atml_recv(struct tpm_chip
 	size = be32_to_cpu(*native_size);
 
 	if (count < size) {
-		dev_err(chip->pdev,
+		dev_err(&chip->dev,
 			"Recv size(%d) less than available space\n", size);
 		for (; i < size; i++) {	/* clear the waiting data anyway */
 			status = ioread8(chip->vendor.iobase + 1);
 			if ((status & ATML_STATUS_DATA_AVAIL) == 0) {
-				dev_err(chip->pdev, "error reading data\n");
+				dev_err(&chip->dev, "error reading data\n");
 				return -EIO;
 			}
 		}

@@ -76,7 +76,7 @@ static int tpm_atml_recv(struct tpm_chip
 	for (; i < size; i++) {
 		status = ioread8(chip->vendor.iobase + 1);
 		if ((status & ATML_STATUS_DATA_AVAIL) == 0) {
-			dev_err(chip->pdev, "error reading data\n");
+			dev_err(&chip->dev, "error reading data\n");
 			return -EIO;
 		}
 		*buf++ = ioread8(chip->vendor.iobase);

@@ -86,7 +86,7 @@ static int tpm_atml_recv(struct tpm_chip
 	status = ioread8(chip->vendor.iobase + 1);
 
 	if (status & ATML_STATUS_DATA_AVAIL) {
-		dev_err(chip->pdev, "data available is stuck\n");
+		dev_err(&chip->dev, "data available is stuck\n");
 		return -EIO;
 	}
 

@@ -97,9 +97,9 @@ static int tpm_atml_send(struct tpm_chip
 {
 	int i;
 
-	dev_dbg(chip->pdev, "tpm_atml_send:\n");
+	dev_dbg(&chip->dev, "tpm_atml_send:\n");
 	for (i = 0; i < count; i++) {
-		dev_dbg(chip->pdev, "%d 0x%x(%d)\n",  i, buf[i], buf[i]);
+		dev_dbg(&chip->dev, "%d 0x%x(%d)\n",  i, buf[i], buf[i]);
  		iowrite8(buf[i], chip->vendor.iobase);
 	}
 

@@ -52,7 +52,7 @@ struct priv_data {
 static int i2c_atmel_send(struct tpm_chip *chip, u8 *buf, size_t len)
 {
 	struct priv_data *priv = chip->vendor.priv;
-	struct i2c_client *client = to_i2c_client(chip->pdev);
+	struct i2c_client *client = to_i2c_client(chip->dev.parent);
 	s32 status;
 
 	priv->len = 0;

@@ -62,7 +62,7 @@ static int i2c_atmel_send(struct tpm_chi
 
 	status = i2c_master_send(client, buf, len);
 
-	dev_dbg(chip->pdev,
+	dev_dbg(&chip->dev,
 		"%s(buf=%*ph len=%0zx) -> sts=%d\n", __func__,
 		(int)min_t(size_t, 64, len), buf, len, status);
 	return status;

@@ -71,7 +71,7 @@ static int i2c_atmel_send(struct tpm_chi
 static int i2c_atmel_recv(struct tpm_chip *chip, u8 *buf, size_t count)
 {
 	struct priv_data *priv = chip->vendor.priv;
-	struct i2c_client *client = to_i2c_client(chip->pdev);
+	struct i2c_client *client = to_i2c_client(chip->dev.parent);
 	struct tpm_output_header *hdr =
 		(struct tpm_output_header *)priv->buffer;
 	u32 expected_len;

@@ -88,7 +88,7 @@ static int i2c_atmel_recv(struct tpm_chi
 		return -ENOMEM;
 
 	if (priv->len >= expected_len) {
-		dev_dbg(chip->pdev,
+		dev_dbg(&chip->dev,
 			"%s early(buf=%*ph count=%0zx) -> ret=%d\n", __func__,
 			(int)min_t(size_t, 64, expected_len), buf, count,
 			expected_len);

@@ -97,7 +97,7 @@ static int i2c_atmel_recv(struct tpm_chi
 	}
 
 	rc = i2c_master_recv(client, buf, expected_len);
-	dev_dbg(chip->pdev,
+	dev_dbg(&chip->dev,
 		"%s reread(buf=%*ph count=%0zx) -> ret=%d\n", __func__,
 		(int)min_t(size_t, 64, expected_len), buf, count,
 		expected_len);

@@ -106,13 +106,13 @@ static int i2c_atmel_recv(struct tpm_chi
 
 static void i2c_atmel_cancel(struct tpm_chip *chip)
 {
-	dev_err(chip->pdev, "TPM operation cancellation was requested, but is not supported");
+	dev_err(&chip->dev, "TPM operation cancellation was requested, but is not supported");
 }
 
 static u8 i2c_atmel_read_status(struct tpm_chip *chip)
 {
 	struct priv_data *priv = chip->vendor.priv;
-	struct i2c_client *client = to_i2c_client(chip->pdev);
+	struct i2c_client *client = to_i2c_client(chip->dev.parent);
 	int rc;
 
 	/* The TPM fails the I2C read until it is ready, so we do the entire

@@ -125,7 +125,7 @@ static u8 i2c_atmel_read_status(struct t
 	/* Once the TPM has completed the command the command remains readable
 	 * until another command is issued. */
 	rc = i2c_master_recv(client, priv->buffer, sizeof(priv->buffer));
-	dev_dbg(chip->pdev,
+	dev_dbg(&chip->dev,
 		"%s: sts=%d", __func__, rc);
 	if (rc <= 0)
 		return 0;

@@ -446,7 +446,7 @@ static int tpm_tis_i2c_recv(struct tpm_c
 	/* read first 10 bytes, including tag, paramsize, and result */
 	size = recv_data(chip, buf, TPM_HEADER_SIZE);
 	if (size < TPM_HEADER_SIZE) {
-		dev_err(chip->pdev, "Unable to read header\n");
+		dev_err(&chip->dev, "Unable to read header\n");
 		goto out;
 	}
 

@@ -459,14 +459,14 @@ static int tpm_tis_i2c_recv(struct tpm_c
 	size += recv_data(chip, &buf[TPM_HEADER_SIZE],
 			  expected - TPM_HEADER_SIZE);
 	if (size < expected) {
-		dev_err(chip->pdev, "Unable to read remainder of result\n");
+		dev_err(&chip->dev, "Unable to read remainder of result\n");
 		size = -ETIME;
 		goto out;
 	}
 
 	wait_for_stat(chip, TPM_STS_VALID, chip->vendor.timeout_c, &status);
 	if (status & TPM_STS_DATA_AVAIL) {	/* retry? */
-		dev_err(chip->pdev, "Error left over data\n");
+		dev_err(&chip->dev, "Error left over data\n");
 		size = -EIO;
 		goto out;
 	}

@@ -96,13 +96,13 @@ static s32 i2c_nuvoton_write_buf(struct
 /* read TPM_STS register */
 static u8 i2c_nuvoton_read_status(struct tpm_chip *chip)
 {
-	struct i2c_client *client = to_i2c_client(chip->pdev);
+	struct i2c_client *client = to_i2c_client(chip->dev.parent);
 	s32 status;
 	u8 data;
 
 	status = i2c_nuvoton_read_buf(client, TPM_STS, 1, &data);
 	if (status <= 0) {
-		dev_err(chip->pdev, "%s() error return %d\n", __func__,
+		dev_err(&chip->dev, "%s() error return %d\n", __func__,
 			status);
 		data = TPM_STS_ERR_VAL;
 	}

@@ -127,13 +127,13 @@ static s32 i2c_nuvoton_write_status(stru
 /* write commandReady to TPM_STS register */
 static void i2c_nuvoton_ready(struct tpm_chip *chip)
 {
-	struct i2c_client *client = to_i2c_client(chip->pdev);
+	struct i2c_client *client = to_i2c_client(chip->dev.parent);
 	s32 status;
 
 	/* this causes the current command to be aborted */
 	status = i2c_nuvoton_write_status(client, TPM_STS_COMMAND_READY);
 	if (status < 0)
-		dev_err(chip->pdev,
+		dev_err(&chip->dev,
 			"%s() fail to write TPM_STS.commandReady\n", __func__);
 }
 

@@ -212,7 +212,7 @@ static int i2c_nuvoton_wait_for_stat(str
 				return 0;
 		} while (time_before(jiffies, stop));
 	}
-	dev_err(chip->pdev, "%s(%02x, %02x) -> timeout\n", __func__, mask,
+	dev_err(&chip->dev, "%s(%02x, %02x) -> timeout\n", __func__, mask,
 		value);
 	return -ETIMEDOUT;
 }

@@ -240,7 +240,7 @@ static int i2c_nuvoton_recv_data(struct
 					       &chip->vendor.read_queue) == 0) {
 		burst_count = i2c_nuvoton_get_burstcount(client, chip);
 		if (burst_count < 0) {
-			dev_err(chip->pdev,
+			dev_err(&chip->dev,
 				"%s() fail to read burstCount=%d\n", __func__,
 				burst_count);
 			return -EIO;

@@ -249,12 +249,12 @@ static int i2c_nuvoton_recv_data(struct
 		rc = i2c_nuvoton_read_buf(client, TPM_DATA_FIFO_R,
 					  bytes2read, &buf[size]);
 		if (rc < 0) {
-			dev_err(chip->pdev,
+			dev_err(&chip->dev,
 				"%s() fail on i2c_nuvoton_read_buf()=%d\n",
 				__func__, rc);
 			return -EIO;
 		}
-		dev_dbg(chip->pdev, "%s(%d):", __func__, bytes2read);
+		dev_dbg(&chip->dev, "%s(%d):", __func__, bytes2read);
 		size += bytes2read;
 	}
 

@@ -264,7 +264,7 @@ static int i2c_nuvoton_recv_data(struct
 /* Read TPM command results */
 static int i2c_nuvoton_recv(struct tpm_chip *chip, u8 *buf, size_t count)
 {
-	struct device *dev = chip->pdev;
+	struct device *dev = chip->dev.parent;
 	struct i2c_client *client = to_i2c_client(dev);
 	s32 rc;
 	int expected, status, burst_count, retries, size = 0;

@@ -334,7 +334,7 @@ static int i2c_nuvoton_recv(struct tpm_c
 		break;
 	}
 	i2c_nuvoton_ready(chip);
-	dev_dbg(chip->pdev, "%s() -> %d\n", __func__, size);
+	dev_dbg(&chip->dev, "%s() -> %d\n", __func__, size);
 	return size;
 }
 

@@ -347,7 +347,7 @@ static int i2c_nuvoton_recv(struct tpm_c
  */
 static int i2c_nuvoton_send(struct tpm_chip *chip, u8 *buf, size_t len)
 {
-	struct device *dev = chip->pdev;
+	struct device *dev = chip->dev.parent;
 	struct i2c_client *client = to_i2c_client(dev);
 	u32 ordinal;
 	size_t count = 0;

@@ -195,9 +195,9 @@ static int wait(struct tpm_chip *chip, i
 	}
 	if (i == TPM_MAX_TRIES) {	/* timeout occurs */
 		if (wait_for_bit == STAT_XFE)
-			dev_err(chip->pdev, "Timeout in wait(STAT_XFE)\n");
+			dev_err(&chip->dev, "Timeout in wait(STAT_XFE)\n");
 		if (wait_for_bit == STAT_RDA)
-			dev_err(chip->pdev, "Timeout in wait(STAT_RDA)\n");
+			dev_err(&chip->dev, "Timeout in wait(STAT_RDA)\n");
 		return -EIO;
 	}
 	return 0;

@@ -220,7 +220,7 @@ static void wait_and_send(struct tpm_chi
 static void tpm_wtx(struct tpm_chip *chip)
 {
 	number_of_wtx++;
-	dev_info(chip->pdev, "Granting WTX (%02d / %02d)\n",
+	dev_info(&chip->dev, "Granting WTX (%02d / %02d)\n",
 		 number_of_wtx, TPM_MAX_WTX_PACKAGES);
 	wait_and_send(chip, TPM_VL_VER);
 	wait_and_send(chip, TPM_CTRL_WTX);

@@ -231,7 +231,7 @@ static void tpm_wtx(struct tpm_chip *chi
 
 static void tpm_wtx_abort(struct tpm_chip *chip)
 {
-	dev_info(chip->pdev, "Aborting WTX\n");
+	dev_info(&chip->dev, "Aborting WTX\n");
 	wait_and_send(chip, TPM_VL_VER);
 	wait_and_send(chip, TPM_CTRL_WTX_ABORT);
 	wait_and_send(chip, 0x00);

@@ -257,7 +257,7 @@ recv_begin:
 	}
 
 	if (buf[0] != TPM_VL_VER) {
-		dev_err(chip->pdev,
+		dev_err(&chip->dev,
 			"Wrong transport protocol implementation!\n");
 		return -EIO;
 	}

@@ -272,7 +272,7 @@ recv_begin:
 		}
 
 		if ((size == 0x6D00) && (buf[1] == 0x80)) {
-			dev_err(chip->pdev, "Error handling on vendor layer!\n");
+			dev_err(&chip->dev, "Error handling on vendor layer!\n");
 			return -EIO;
 		}
 

@@ -284,7 +284,7 @@ recv_begin:
 	}
 
 	if (buf[1] == TPM_CTRL_WTX) {
-		dev_info(chip->pdev, "WTX-package received\n");
+		dev_info(&chip->dev, "WTX-package received\n");
 		if (number_of_wtx < TPM_MAX_WTX_PACKAGES) {
 			tpm_wtx(chip);
 			goto recv_begin;

@@ -295,14 +295,14 @@ recv_begin:
 	}
 
 	if (buf[1] == TPM_CTRL_WTX_ABORT_ACK) {
-		dev_info(chip->pdev, "WTX-abort acknowledged\n");
+		dev_info(&chip->dev, "WTX-abort acknowledged\n");
 		return size;
 	}
 
 	if (buf[1] == TPM_CTRL_ERROR) {
-		dev_err(chip->pdev, "ERROR-package received:\n");
+		dev_err(&chip->dev, "ERROR-package received:\n");
 		if (buf[4] == TPM_INF_NAK)
-			dev_err(chip->pdev,
+			dev_err(&chip->dev,
 				"-> Negative acknowledgement"
 				" - retransmit command!\n");
 		return -EIO;

@@ -321,7 +321,7 @@ static int tpm_inf_send(struct tpm_chip
 
 	ret = empty_fifo(chip, 1);
 	if (ret) {
-		dev_err(chip->pdev, "Timeout while clearing FIFO\n");
+		dev_err(&chip->dev, "Timeout while clearing FIFO\n");
 		return -EIO;
 	}
 

@@ -113,7 +113,7 @@ static int nsc_wait_for_ready(struct tpm
 	}
 	while (time_before(jiffies, stop));
 
-	dev_info(chip->pdev, "wait for ready failed\n");
+	dev_info(&chip->dev, "wait for ready failed\n");
 	return -EBUSY;
 }
 

@@ -129,12 +129,12 @@ static int tpm_nsc_recv(struct tpm_chip
 		return -EIO;
 
 	if (wait_for_stat(chip, NSC_STATUS_F0, NSC_STATUS_F0, &data) < 0) {
-		dev_err(chip->pdev, "F0 timeout\n");
+		dev_err(&chip->dev, "F0 timeout\n");
 		return -EIO;
 	}
 	if ((data =
 	     inb(chip->vendor.base + NSC_DATA)) != NSC_COMMAND_NORMAL) {
-		dev_err(chip->pdev, "not in normal mode (0x%x)\n",
+		dev_err(&chip->dev, "not in normal mode (0x%x)\n",
 			data);
 		return -EIO;
 	}

@@ -143,7 +143,7 @@ static int tpm_nsc_recv(struct tpm_chip
 	for (p = buffer; p < &buffer[count]; p++) {
 		if (wait_for_stat
 		    (chip, NSC_STATUS_OBF, NSC_STATUS_OBF, &data) < 0) {
-			dev_err(chip->pdev,
+			dev_err(&chip->dev,
 				"OBF timeout (while reading data)\n");
 			return -EIO;
 		}

@@ -154,11 +154,11 @@ static int tpm_nsc_recv(struct tpm_chip
 
 	if ((data & NSC_STATUS_F0) == 0 &&
 	(wait_for_stat(chip, NSC_STATUS_F0, NSC_STATUS_F0, &data) < 0)) {
-		dev_err(chip->pdev, "F0 not set\n");
+		dev_err(&chip->dev, "F0 not set\n");
 		return -EIO;
 	}
 	if ((data = inb(chip->vendor.base + NSC_DATA)) != NSC_COMMAND_EOC) {
-		dev_err(chip->pdev,
+		dev_err(&chip->dev,
 			"expected end of command(0x%x)\n", data);
 		return -EIO;
 	}

@@ -189,19 +189,19 @@ static int tpm_nsc_send(struct tpm_chip
 		return -EIO;
 
 	if (wait_for_stat(chip, NSC_STATUS_IBF, 0, &data) < 0) {
-		dev_err(chip->pdev, "IBF timeout\n");
+		dev_err(&chip->dev, "IBF timeout\n");
 		return -EIO;
 	}
 
 	outb(NSC_COMMAND_NORMAL, chip->vendor.base + NSC_COMMAND);
 	if (wait_for_stat(chip, NSC_STATUS_IBR, NSC_STATUS_IBR, &data) < 0) {
-		dev_err(chip->pdev, "IBR timeout\n");
+		dev_err(&chip->dev, "IBR timeout\n");
 		return -EIO;
 	}
 
 	for (i = 0; i < count; i++) {
 		if (wait_for_stat(chip, NSC_STATUS_IBF, 0, &data) < 0) {
-			dev_err(chip->pdev,
+			dev_err(&chip->dev,
 				"IBF timeout (while writing data)\n");
 			return -EIO;
 		}

@@ -209,7 +209,7 @@ static int tpm_nsc_send(struct tpm_chip
 	}
 
 	if (wait_for_stat(chip, NSC_STATUS_IBF, 0, &data) < 0) {
-		dev_err(chip->pdev, "IBF timeout\n");
+		dev_err(&chip->dev, "IBF timeout\n");
 		return -EIO;
 	}
 	outb(NSC_COMMAND_EOC, chip->vendor.base + NSC_COMMAND);

@@ -293,7 +293,7 @@ static int tpm_tis_recv(struct tpm_chip
 	/* read first 10 bytes, including tag, paramsize, and result */
 	if ((size =
 	     recv_data(chip, buf, TPM_HEADER_SIZE)) < TPM_HEADER_SIZE) {
-		dev_err(chip->pdev, "Unable to read header\n");
+		dev_err(&chip->dev, "Unable to read header\n");
 		goto out;
 	}
 

@@ -306,7 +306,7 @@ static int tpm_tis_recv(struct tpm_chip
 	if ((size +=
 	     recv_data(chip, &buf[TPM_HEADER_SIZE],
 		       expected - TPM_HEADER_SIZE)) < expected) {
-		dev_err(chip->pdev, "Unable to read remainder of result\n");
+		dev_err(&chip->dev, "Unable to read remainder of result\n");
 		size = -ETIME;
 		goto out;
 	}

@@ -315,7 +315,7 @@ static int tpm_tis_recv(struct tpm_chip
 			  &chip->vendor.int_queue, false);
 	status = tpm_tis_status(chip);
 	if (status & TPM_STS_DATA_AVAIL) {	/* retry? */
-		dev_err(chip->pdev, "Error left over data\n");
+		dev_err(&chip->dev, "Error left over data\n");
 		size = -EIO;
 		goto out;
 	}

@@ -401,7 +401,7 @@ static void disable_interrupts(struct tp
 	iowrite32(intmask,
 		  chip->vendor.iobase +
 		  TPM_INT_ENABLE(chip->vendor.locality));
-	devm_free_irq(chip->pdev, chip->vendor.irq, chip);
+	devm_free_irq(&chip->dev, chip->vendor.irq, chip);
 	chip->vendor.irq = 0;
 }
 

@@ -463,7 +463,7 @@ static int tpm_tis_send(struct tpm_chip
 		msleep(1);
 	if (!priv->irq_tested) {
 		disable_interrupts(chip);
-		dev_err(chip->pdev,
+		dev_err(&chip->dev,
 			FW_BUG "TPM interrupt not working, polling instead\n");
 	}
 	priv->irq_tested = true;

@@ -533,7 +533,7 @@ static int probe_itpm(struct tpm_chip *c
 
 	rc = tpm_tis_send_data(chip, cmd_getticks, len);
 	if (rc == 0) {
-		dev_info(chip->pdev, "Detected an iTPM.\n");
+		dev_info(&chip->dev, "Detected an iTPM.\n");
 		rc = 1;
 	} else
 		rc = -EFAULT;

@@ -766,7 +766,7 @@ static int tpm_tis_init(struct device *d
 			if (devm_request_irq
 			    (dev, i, tis_int_probe, IRQF_SHARED,
 			     chip->devname, chip) != 0) {
-				dev_info(chip->pdev,
+				dev_info(&chip->dev,
 					 "Unable to request irq: %d for probe\n",
 					 i);
 				continue;

@@ -818,7 +818,7 @@ static int tpm_tis_init(struct device *d
 		if (devm_request_irq
 		    (dev, chip->vendor.irq, tis_int_handler, IRQF_SHARED,
 		     chip->devname, chip) != 0) {
-			dev_info(chip->pdev,
+			dev_info(&chip->dev,
 				 "Unable to request irq: %d for use\n",
 				 chip->vendor.irq);
 			chip->vendor.irq = 0;

@@ -36,10 +36,60 @@ static DEFINE_SPINLOCK(driver_lock);
 struct class *tpm_class;
 dev_t tpm_devt;
 
-/*
- * tpm_chip_find_get - return tpm_chip for a given chip number
- * @chip_num the device number for the chip
+/**
+ * tpm_try_get_ops() - Get a ref to the tpm_chip
+ * @chip: Chip to ref
+ *
+ * The caller must already have some kind of locking to ensure that chip is
+ * valid. This function will lock the chip so that the ops member can be
+ * accessed safely. The locking prevents tpm_chip_unregister from
+ * completing, so it should not be held for long periods.
+ *
+ * Returns -ERRNO if the chip could not be got.
  */
+int tpm_try_get_ops(struct tpm_chip *chip)
+{
+	int rc = -EIO;
+
+	get_device(&chip->dev);
+
+	down_read(&chip->ops_sem);
+	if (!chip->ops)
+		goto out_lock;
+
+	if (!try_module_get(chip->dev.parent->driver->owner))
+		goto out_lock;
+
+	return 0;
+out_lock:
+	up_read(&chip->ops_sem);
+	put_device(&chip->dev);
+	return rc;
+}
+EXPORT_SYMBOL_GPL(tpm_try_get_ops);
+
+/**
+ * tpm_put_ops() - Release a ref to the tpm_chip
+ * @chip: Chip to put
+ *
+ * This is the opposite pair to tpm_try_get_ops(). After this returns chip may
+ * be kfree'd.
+ */
+void tpm_put_ops(struct tpm_chip *chip)
+{
+	module_put(chip->dev.parent->driver->owner);
+	up_read(&chip->ops_sem);
+	put_device(&chip->dev);
+}
+EXPORT_SYMBOL_GPL(tpm_put_ops);
+
+/**
+ * tpm_chip_find_get() - return tpm_chip for a given chip number
+ * @chip_num: id to find
+ *
+ * The return'd chip has been tpm_try_get_ops'd and must be released via
+ * tpm_put_ops
+  */
 struct tpm_chip *tpm_chip_find_get(int chip_num)
 {
 	struct tpm_chip *pos, *chip = NULL;

@@ -49,10 +99,10 @@ struct tpm_chip *tpm_chip_find_get(int c
 		if (chip_num != TPM_ANY_NUM && chip_num != pos->dev_num)
 			continue;
 
-		if (try_module_get(pos->dev.parent->driver->owner)) {
+		/* rcu prevents chip from being free'd */
+		if (!tpm_try_get_ops(pos))
 			chip = pos;
-			break;
-		}
+		break;
 	}
 	rcu_read_unlock();
 	return chip;

@@ -94,6 +144,7 @@ struct tpm_chip *tpmm_chip_alloc(struct
 		return ERR_PTR(-ENOMEM);
 
 	mutex_init(&chip->tpm_mutex);
+	init_rwsem(&chip->ops_sem);
 	INIT_LIST_HEAD(&chip->list);
 
 	chip->ops = ops;

@@ -171,6 +222,12 @@ static int tpm_add_char_device(struct tp
 static void tpm_del_char_device(struct tpm_chip *chip)
 {
 	cdev_del(&chip->cdev);
+
+	/* Make the driver uncallable. */
+	down_write(&chip->ops_sem);
+	chip->ops = NULL;
+	up_write(&chip->ops_sem);
+
 	device_del(&chip->dev);
 }
 

@@ -256,6 +313,9 @@ EXPORT_SYMBOL_GPL(tpm_chip_register);
  * Takes the chip first away from the list of available TPM chips and then
  * cleans up all the resources reserved by tpm_chip_register().
  *
+ * Once this function returns the driver call backs in 'op's will not be
+ * running and will no longer start.
+ *
  * NOTE: This function should be only called before deinitializing chip
  * resources.
  */

@@ -136,9 +136,18 @@ static ssize_t tpm_write(struct file *fi
 		return -EFAULT;
 	}
 
-	/* atomic tpm command send and result receive */
+	/* atomic tpm command send and result receive. We only hold the ops
+	 * lock during this period so that the tpm can be unregistered even if
+	 * the char dev is held open.
+	 */
+	if (tpm_try_get_ops(priv->chip)) {
+		mutex_unlock(&priv->buffer_mutex);
+		return -EPIPE;
+	}
 	out_size = tpm_transmit(priv->chip, priv->data_buffer,
 				sizeof(priv->data_buffer), 0);
+
+	tpm_put_ops(priv->chip);
 	if (out_size < 0) {
 		mutex_unlock(&priv->buffer_mutex);
 		return out_size;

@@ -687,7 +687,7 @@ int tpm_is_tpm2(u32 chip_num)
 
 	rc = (chip->flags & TPM_CHIP_FLAG_TPM2) != 0;
 
-	tpm_chip_put(chip);
+	tpm_put_ops(chip);
 
 	return rc;
 }

@@ -716,7 +716,7 @@ int tpm_pcr_read(u32 chip_num, int pcr_i
 		rc = tpm2_pcr_read(chip, pcr_idx, res_buf);
 	else
 		rc = tpm_pcr_read_dev(chip, pcr_idx, res_buf);
-	tpm_chip_put(chip);
+	tpm_put_ops(chip);
 	return rc;
 }
 EXPORT_SYMBOL_GPL(tpm_pcr_read);

@@ -751,7 +751,7 @@ int tpm_pcr_extend(u32 chip_num, int pcr
 
 	if (chip->flags & TPM_CHIP_FLAG_TPM2) {
 		rc = tpm2_pcr_extend(chip, pcr_idx, hash);
-		tpm_chip_put(chip);
+		tpm_put_ops(chip);
 		return rc;
 	}
 

@@ -761,7 +761,7 @@ int tpm_pcr_extend(u32 chip_num, int pcr
 	rc = tpm_transmit_cmd(chip, &cmd, EXTEND_PCR_RESULT_SIZE, 0,
 			      "attempting extend a PCR value");
 
-	tpm_chip_put(chip);
+	tpm_put_ops(chip);
 	return rc;
 }
 EXPORT_SYMBOL_GPL(tpm_pcr_extend);

@@ -842,7 +842,7 @@ int tpm_send(u32 chip_num, void *cmd, si
 
 	rc = tpm_transmit_cmd(chip, cmd, buflen, 0, "attempting tpm_cmd");
 
-	tpm_chip_put(chip);
+	tpm_put_ops(chip);
 	return rc;
 }
 EXPORT_SYMBOL_GPL(tpm_send);

@@ -1025,7 +1025,7 @@ int tpm_get_random(u32 chip_num, u8 *out
 
 	if (chip->flags & TPM_CHIP_FLAG_TPM2) {
 		err = tpm2_get_random(chip, out, max);
-		tpm_chip_put(chip);
+		tpm_put_ops(chip);
 		return err;
 	}
 

@@ -1047,7 +1047,7 @@ int tpm_get_random(u32 chip_num, u8 *out
 		num_bytes -= recd;
 	} while (retries-- && total < max);
 
-	tpm_chip_put(chip);
+	tpm_put_ops(chip);
 	return total ? total : -EIO;
 }
 EXPORT_SYMBOL_GPL(tpm_get_random);

@@ -1073,7 +1073,7 @@ int tpm_seal_trusted(u32 chip_num, struc
 
 	rc = tpm2_seal_trusted(chip, payload, options);
 
-	tpm_chip_put(chip);
+	tpm_put_ops(chip);
 	return rc;
 }
 EXPORT_SYMBOL_GPL(tpm_seal_trusted);

@@ -1099,7 +1099,8 @@ int tpm_unseal_trusted(u32 chip_num, str
 
 	rc = tpm2_unseal_trusted(chip, payload, options);
 
-	tpm_chip_put(chip);
+	tpm_put_ops(chip);
+
 	return rc;
 }
 EXPORT_SYMBOL_GPL(tpm_unseal_trusted);

@@ -295,5 +295,10 @@ int tpm_sysfs_add_device(struct tpm_chip
 
 void tpm_sysfs_del_device(struct tpm_chip *chip)
 {
+	/* The sysfs routines rely on an implicit tpm_try_get_ops, this
+	 * function is called before ops is null'd and the sysfs core
+	 * synchronizes this removal so that no callbacks are running or can
+	 * run again
+	 */
 	sysfs_remove_group(&chip->dev.parent->kobj, &tpm_dev_group);
 }

@@ -174,7 +174,13 @@ struct tpm_chip {
 	struct device dev;
 	struct cdev cdev;
 
+	/* A driver callback under ops cannot be run unless ops_sem is held
+	 * (sometimes implicitly, eg for the sysfs code). ops becomes null
+	 * when the driver is unregistered, see tpm_try_get_ops.
+	 */
+	struct rw_semaphore ops_sem;
 	const struct tpm_class_ops *ops;
+
 	unsigned int flags;
 
 	int dev_num;		/* /dev/tpm# */

@@ -200,11 +206,6 @@ struct tpm_chip {
 
 #define to_tpm_chip(d) container_of(d, struct tpm_chip, dev)
 
-static inline void tpm_chip_put(struct tpm_chip *chip)
-{
-	module_put(chip->dev.parent->driver->owner);
-}
-
 static inline int tpm_read_index(int base, int index)
 {
 	outb(index, base);

@@ -516,6 +517,9 @@ extern int wait_for_tpm_stat(struct tpm_
 			     wait_queue_head_t *, bool);
 
 struct tpm_chip *tpm_chip_find_get(int chip_num);
+__must_check int tpm_try_get_ops(struct tpm_chip *chip);
+void tpm_put_ops(struct tpm_chip *chip);
+
 extern struct tpm_chip *tpmm_chip_alloc(struct device *dev,
 				       const struct tpm_class_ops *ops);
 extern int tpm_chip_register(struct tpm_chip *chip);