Merge tag v3.10.44 into linux-linaro-lsk

This is the 3.10.44 stable release.

Makefile

@@ -1,6 +1,6 @@
 VERSION = 3
 PATCHLEVEL = 10
-SUBLEVEL = 43
+SUBLEVEL = 44
 EXTRAVERSION =
 NAME = TOSSUG Baby Fish
 

arch/arm/boot/dts/armada-xp-gp.dts

@@ -124,7 +124,7 @@
 				/* Device Bus parameters are required */
 
 				/* Read parameters */
-				devbus,bus-width    = <8>;
+				devbus,bus-width    = <16>;
 				devbus,turn-off-ps  = <60000>;
 				devbus,badr-skew-ps = <0>;
 				devbus,acc-first-ps = <124000>;

arch/arm/boot/dts/armada-xp-openblocks-ax3-4.dts

@@ -152,7 +152,7 @@
 				/* Device Bus parameters are required */
 
 				/* Read parameters */
-				devbus,bus-width    = <8>;
+				devbus,bus-width    = <16>;
 				devbus,turn-off-ps  = <60000>;
 				devbus,badr-skew-ps = <0>;
 				devbus,acc-first-ps = <124000>;

drivers/ata/ahci.c

@@ -444,10 +444,14 @@ static const struct pci_device_id ahci_pci_tbl[] = {
 	  .driver_data = board_ahci_yes_fbs },			/* 88se9172 */
 	{ PCI_DEVICE(PCI_VENDOR_ID_MARVELL_EXT, 0x9192),
 	  .driver_data = board_ahci_yes_fbs },			/* 88se9172 on some Gigabyte */
+	{ PCI_DEVICE(PCI_VENDOR_ID_MARVELL_EXT, 0x91a0),
+	  .driver_data = board_ahci_yes_fbs },
 	{ PCI_DEVICE(PCI_VENDOR_ID_MARVELL_EXT, 0x91a3),
 	  .driver_data = board_ahci_yes_fbs },
 	{ PCI_DEVICE(PCI_VENDOR_ID_MARVELL_EXT, 0x9230),
 	  .driver_data = board_ahci_yes_fbs },
+	{ PCI_DEVICE(PCI_VENDOR_ID_TTI, 0x0642),
+	  .driver_data = board_ahci_yes_fbs },
 
 	/* Promise */
 	{ PCI_VDEVICE(PROMISE, 0x3f20), board_ahci },	/* PDC42819 */

drivers/infiniband/ulp/isert/ib_isert.c

@@ -382,6 +382,14 @@ isert_connect_request(struct rdma_cm_id *cma_id, struct rdma_cm_event *event)
 	struct ib_device *ib_dev = cma_id->device;
 	int ret = 0;
 
+	spin_lock_bh(&np->np_thread_lock);
+	if (!np->enabled) {
+		spin_unlock_bh(&np->np_thread_lock);
+		pr_debug("iscsi_np is not enabled, reject connect request\n");
+		return rdma_reject(cma_id, NULL, 0);
+	}
+	spin_unlock_bh(&np->np_thread_lock);
+
 	pr_debug("Entering isert_connect_request cma_id: %p, context: %p\n",
 		 cma_id, cma_id->context);
 

drivers/misc/mei/hw-me.c

@@ -183,6 +183,7 @@ static void mei_me_hw_reset(struct mei_device *dev, bool intr_enable)
 	else
 		hcsr &= ~H_IE;
 
+	dev->recvd_hw_ready = false;
 	mei_me_reg_write(hw, H_CSR, hcsr);
 
 	if (dev->dev_state == MEI_DEV_POWER_DOWN)
@@ -233,10 +234,7 @@ static bool mei_me_hw_is_ready(struct mei_device *dev)
 static int mei_me_hw_ready_wait(struct mei_device *dev)
 {
 	int err;
-	if (mei_me_hw_is_ready(dev))
-		return 0;
 
-	dev->recvd_hw_ready = false;
 	mutex_unlock(&dev->device_lock);
 	err = wait_event_interruptible_timeout(dev->wait_hw_ready,
 			dev->recvd_hw_ready,

drivers/net/ethernet/mellanox/mlx4/en_cq.c

@@ -55,7 +55,6 @@ int mlx4_en_create_cq(struct mlx4_en_priv *priv,
 
 	cq->ring = ring;
 	cq->is_tx = mode;
-	spin_lock_init(&cq->lock);
 
 	err = mlx4_alloc_hwq_res(mdev->dev, &cq->wqres,
 				cq->buf_size, 2 * PAGE_SIZE);

drivers/net/ethernet/mellanox/mlx4/en_netdev.c

@@ -1190,15 +1190,11 @@ static void mlx4_en_netpoll(struct net_device *dev)
 {
 	struct mlx4_en_priv *priv = netdev_priv(dev);
 	struct mlx4_en_cq *cq;
-	unsigned long flags;
 	int i;
 
 	for (i = 0; i < priv->rx_ring_num; i++) {
 		cq = &priv->rx_cq[i];
-		spin_lock_irqsave(&cq->lock, flags);
-		napi_synchronize(&cq->napi);
-		mlx4_en_process_rx_cq(dev, cq, 0);
-		spin_unlock_irqrestore(&cq->lock, flags);
+		napi_schedule(&cq->napi);
 	}
 }
 #endif

drivers/net/ethernet/mellanox/mlx4/mlx4_en.h

@@ -299,7 +299,6 @@ struct mlx4_en_cq {
 	struct mlx4_cq          mcq;
 	struct mlx4_hwq_resources wqres;
 	int                     ring;
-	spinlock_t              lock;
 	struct net_device      *dev;
 	struct napi_struct	napi;
 	int size;

drivers/scsi/megaraid/megaraid_sas.h

@@ -1295,7 +1295,6 @@ struct megasas_instance {
 	u32 *reply_queue;
 	dma_addr_t reply_queue_h;
 
-	unsigned long base_addr;
 	struct megasas_register_set __iomem *reg_set;
 
 	struct megasas_pd_list          pd_list[MEGASAS_MAX_PD];

drivers/scsi/megaraid/megaraid_sas_base.c

@@ -3461,6 +3461,7 @@ static int megasas_init_fw(struct megasas_instance *instance)
 	u32 max_sectors_1;
 	u32 max_sectors_2;
 	u32 tmp_sectors, msix_enable;
+	resource_size_t base_addr;
 	struct megasas_register_set __iomem *reg_set;
 	struct megasas_ctrl_info *ctrl_info;
 	unsigned long bar_list;
@@ -3469,14 +3470,14 @@ static int megasas_init_fw(struct megasas_instance *instance)
 	/* Find first memory bar */
 	bar_list = pci_select_bars(instance->pdev, IORESOURCE_MEM);
 	instance->bar = find_first_bit(&bar_list, sizeof(unsigned long));
-	instance->base_addr = pci_resource_start(instance->pdev, instance->bar);
 	if (pci_request_selected_regions(instance->pdev, instance->bar,
 					 "megasas: LSI")) {
 		printk(KERN_DEBUG "megasas: IO memory region busy!\n");
 		return -EBUSY;
 	}
 
-	instance->reg_set = ioremap_nocache(instance->base_addr, 8192);
+	base_addr = pci_resource_start(instance->pdev, instance->bar);
+	instance->reg_set = ioremap_nocache(base_addr, 8192);
 
 	if (!instance->reg_set) {
 		printk(KERN_DEBUG "megasas: Failed to map IO mem\n");

drivers/target/iscsi/iscsi_target.c

@@ -460,6 +460,7 @@ int iscsit_del_np(struct iscsi_np *np)
 	spin_lock_bh(&np->np_thread_lock);
 	np->np_exports--;
 	if (np->np_exports) {
+		np->enabled = true;
 		spin_unlock_bh(&np->np_thread_lock);
 		return 0;
 	}

drivers/target/iscsi/iscsi_target_core.h

@@ -760,6 +760,7 @@ struct iscsi_np {
 	int			np_ip_proto;
 	int			np_sock_type;
 	enum np_thread_state_table np_thread_state;
+	bool                    enabled;
 	enum iscsi_timer_flags_table np_login_timer_flags;
 	u32			np_exports;
 	enum np_flags_table	np_flags;

drivers/target/iscsi/iscsi_target_login.c

@@ -250,6 +250,28 @@ static void iscsi_login_set_conn_values(
 	mutex_unlock(&auth_id_lock);
 }
 
+static __printf(2, 3) int iscsi_change_param_sprintf(
+	struct iscsi_conn *conn,
+	const char *fmt, ...)
+{
+	va_list args;
+	unsigned char buf[64];
+
+	memset(buf, 0, sizeof buf);
+
+	va_start(args, fmt);
+	vsnprintf(buf, sizeof buf, fmt, args);
+	va_end(args);
+
+	if (iscsi_change_param_value(buf, conn->param_list, 0) < 0) {
+		iscsit_tx_login_rsp(conn, ISCSI_STATUS_CLS_TARGET_ERR,
+				ISCSI_LOGIN_STATUS_NO_RESOURCES);
+		return -1;
+	}
+
+	return 0;
+}
+
 /*
  *	This is the leading connection of a new session,
  *	or session reinstatement.
@@ -339,7 +361,6 @@ static int iscsi_login_zero_tsih_s2(
 {
 	struct iscsi_node_attrib *na;
 	struct iscsi_session *sess = conn->sess;
-	unsigned char buf[32];
 	bool iser = false;
 
 	sess->tpg = conn->tpg;
@@ -380,26 +401,16 @@ static int iscsi_login_zero_tsih_s2(
 	 *
 	 * In our case, we have already located the struct iscsi_tiqn at this point.
 	 */
-	memset(buf, 0, 32);
-	sprintf(buf, "TargetPortalGroupTag=%hu", ISCSI_TPG_S(sess)->tpgt);
-	if (iscsi_change_param_value(buf, conn->param_list, 0) < 0) {
-		iscsit_tx_login_rsp(conn, ISCSI_STATUS_CLS_TARGET_ERR,
-				ISCSI_LOGIN_STATUS_NO_RESOURCES);
+	if (iscsi_change_param_sprintf(conn, "TargetPortalGroupTag=%hu", sess->tpg->tpgt))
 		return -1;
-	}
 
 	/*
 	 * Workaround for Initiators that have broken connection recovery logic.
 	 *
 	 * "We would really like to get rid of this." Linux-iSCSI.org team
 	 */
-	memset(buf, 0, 32);
-	sprintf(buf, "ErrorRecoveryLevel=%d", na->default_erl);
-	if (iscsi_change_param_value(buf, conn->param_list, 0) < 0) {
-		iscsit_tx_login_rsp(conn, ISCSI_STATUS_CLS_TARGET_ERR,
-				ISCSI_LOGIN_STATUS_NO_RESOURCES);
+	if (iscsi_change_param_sprintf(conn, "ErrorRecoveryLevel=%d", na->default_erl))
 		return -1;
-	}
 
 	if (iscsi_login_disable_FIM_keys(conn->param_list, conn) < 0)
 		return -1;
@@ -411,12 +422,9 @@ static int iscsi_login_zero_tsih_s2(
 		unsigned long mrdsl, off;
 		int rc;
 
-		sprintf(buf, "RDMAExtensions=Yes");
-		if (iscsi_change_param_value(buf, conn->param_list, 0) < 0) {
-			iscsit_tx_login_rsp(conn, ISCSI_STATUS_CLS_TARGET_ERR,
-				ISCSI_LOGIN_STATUS_NO_RESOURCES);
+		if (iscsi_change_param_sprintf(conn, "RDMAExtensions=Yes"))
 			return -1;
-		}
+
 		/*
 		 * Make MaxRecvDataSegmentLength PAGE_SIZE aligned for
 		 * Immediate Data + Unsolicitied Data-OUT if necessary..
@@ -446,12 +454,8 @@ static int iscsi_login_zero_tsih_s2(
 		pr_warn("Aligning ISER MaxRecvDataSegmentLength: %lu down"
 			" to PAGE_SIZE\n", mrdsl);
 
-		sprintf(buf, "MaxRecvDataSegmentLength=%lu\n", mrdsl);
-		if (iscsi_change_param_value(buf, conn->param_list, 0) < 0) {
-			iscsit_tx_login_rsp(conn, ISCSI_STATUS_CLS_TARGET_ERR,
-				ISCSI_LOGIN_STATUS_NO_RESOURCES);
+		if (iscsi_change_param_sprintf(conn, "MaxRecvDataSegmentLength=%lu\n", mrdsl))
 			return -1;
-		}
 	}
 
 	return 0;
@@ -984,6 +988,7 @@ int iscsi_target_setup_login_socket(
 	}
 
 	np->np_transport = t;
+	np->enabled = true;
 	return 0;
 }
 

drivers/target/iscsi/iscsi_target_tpg.c

@@ -175,13 +175,16 @@ void iscsit_put_tpg(struct iscsi_portal_group *tpg)
 
 static void iscsit_clear_tpg_np_login_thread(
 	struct iscsi_tpg_np *tpg_np,
-	struct iscsi_portal_group *tpg)
+	struct iscsi_portal_group *tpg,
+	bool shutdown)
 {
 	if (!tpg_np->tpg_np) {
 		pr_err("struct iscsi_tpg_np->tpg_np is NULL!\n");
 		return;
 	}
 
+	if (shutdown)
+		tpg_np->tpg_np->enabled = false;
 	iscsit_reset_np_thread(tpg_np->tpg_np, tpg_np, tpg);
 }
 
@@ -197,7 +200,7 @@ void iscsit_clear_tpg_np_login_threads(
 			continue;
 		}
 		spin_unlock(&tpg->tpg_np_lock);
-		iscsit_clear_tpg_np_login_thread(tpg_np, tpg);
+		iscsit_clear_tpg_np_login_thread(tpg_np, tpg, false);
 		spin_lock(&tpg->tpg_np_lock);
 	}
 	spin_unlock(&tpg->tpg_np_lock);
@@ -520,7 +523,7 @@ static int iscsit_tpg_release_np(
 	struct iscsi_portal_group *tpg,
 	struct iscsi_np *np)
 {
-	iscsit_clear_tpg_np_login_thread(tpg_np, tpg);
+	iscsit_clear_tpg_np_login_thread(tpg_np, tpg, true);
 
 	pr_debug("CORE[%s] - Removed Network Portal: %s:%hu,%hu on %s\n",
 		tpg->tpg_tiqn->tiqn, np->np_ip, np->np_port, tpg->tpgt,

drivers/target/target_core_alua.c

@@ -409,7 +409,16 @@ static inline int core_alua_state_standby(
 	case REPORT_LUNS:
 	case RECEIVE_DIAGNOSTIC:
 	case SEND_DIAGNOSTIC:
+	case READ_CAPACITY:
 		return 0;
+	case SERVICE_ACTION_IN:
+		switch (cdb[1] & 0x1f) {
+		case SAI_READ_CAPACITY_16:
+			return 0;
+		default:
+			*alua_ascq = ASCQ_04H_ALUA_TG_PT_STANDBY;
+			return 1;
+		}
 	case MAINTENANCE_IN:
 		switch (cdb[1] & 0x1f) {
 		case MI_REPORT_TARGET_PGS:

drivers/target/target_core_configfs.c

@@ -2034,6 +2034,11 @@ static ssize_t target_core_alua_tg_pt_gp_store_attr_alua_access_state(
 			" tg_pt_gp ID: %hu\n", tg_pt_gp->tg_pt_gp_valid_id);
 		return -EINVAL;
 	}
+	if (!(dev->dev_flags & DF_CONFIGURED)) {
+		pr_err("Unable to set alua_access_state while device is"
+		       " not configured\n");
+		return -ENODEV;
+	}
 
 	ret = strict_strtoul(page, 0, &tmp);
 	if (ret < 0) {

fs/attr.c

@@ -50,14 +50,14 @@ int inode_change_ok(const struct inode *inode, struct iattr *attr)
 	if ((ia_valid & ATTR_UID) &&
 	    (!uid_eq(current_fsuid(), inode->i_uid) ||
 	     !uid_eq(attr->ia_uid, inode->i_uid)) &&
-	    !inode_capable(inode, CAP_CHOWN))
+	    !capable_wrt_inode_uidgid(inode, CAP_CHOWN))
 		return -EPERM;
 
 	/* Make sure caller can chgrp. */
 	if ((ia_valid & ATTR_GID) &&
 	    (!uid_eq(current_fsuid(), inode->i_uid) ||
 	    (!in_group_p(attr->ia_gid) && !gid_eq(attr->ia_gid, inode->i_gid))) &&
-	    !inode_capable(inode, CAP_CHOWN))
+	    !capable_wrt_inode_uidgid(inode, CAP_CHOWN))
 		return -EPERM;
 
 	/* Make sure a caller can chmod. */
@@ -67,7 +67,7 @@ int inode_change_ok(const struct inode *inode, struct iattr *attr)
 		/* Also check the setgid bit! */
 		if (!in_group_p((ia_valid & ATTR_GID) ? attr->ia_gid :
 				inode->i_gid) &&
-		    !inode_capable(inode, CAP_FSETID))
+		    !capable_wrt_inode_uidgid(inode, CAP_FSETID))
 			attr->ia_mode &= ~S_ISGID;
 	}
 
@@ -160,7 +160,7 @@ void setattr_copy(struct inode *inode, const struct iattr *attr)
 		umode_t mode = attr->ia_mode;
 
 		if (!in_group_p(inode->i_gid) &&
-		    !inode_capable(inode, CAP_FSETID))
+		    !capable_wrt_inode_uidgid(inode, CAP_FSETID))
 			mode &= ~S_ISGID;
 		inode->i_mode = mode;
 	}

fs/inode.c

@@ -1837,14 +1837,18 @@ EXPORT_SYMBOL(inode_init_owner);
  * inode_owner_or_capable - check current task permissions to inode
  * @inode: inode being checked
  *
- * Return true if current either has CAP_FOWNER to the inode, or
- * owns the file.
+ * Return true if current either has CAP_FOWNER in a namespace with the
+ * inode owner uid mapped, or owns the file.
  */
 bool inode_owner_or_capable(const struct inode *inode)
 {
+	struct user_namespace *ns;
+
 	if (uid_eq(current_fsuid(), inode->i_uid))
 		return true;
-	if (inode_capable(inode, CAP_FOWNER))
+
+	ns = current_user_ns();
+	if (ns_capable(ns, CAP_FOWNER) && kuid_has_mapping(ns, inode->i_uid))
 		return true;
 	return false;
 }

fs/namei.c

@@ -321,10 +321,11 @@ int generic_permission(struct inode *inode, int mask)
 
 	if (S_ISDIR(inode->i_mode)) {
 		/* DACs are overridable for directories */
-		if (inode_capable(inode, CAP_DAC_OVERRIDE))
+		if (capable_wrt_inode_uidgid(inode, CAP_DAC_OVERRIDE))
 			return 0;
 		if (!(mask & MAY_WRITE))
-			if (inode_capable(inode, CAP_DAC_READ_SEARCH))
+			if (capable_wrt_inode_uidgid(inode,
+						     CAP_DAC_READ_SEARCH))
 				return 0;
 		return -EACCES;
 	}
@@ -334,7 +335,7 @@ int generic_permission(struct inode *inode, int mask)
 	 * at least one exec bit set.
 	 */
 	if (!(mask & MAY_EXEC) || (inode->i_mode & S_IXUGO))
-		if (inode_capable(inode, CAP_DAC_OVERRIDE))
+		if (capable_wrt_inode_uidgid(inode, CAP_DAC_OVERRIDE))
 			return 0;
 
 	/*
@@ -342,7 +343,7 @@ int generic_permission(struct inode *inode, int mask)
 	 */
 	mask &= MAY_READ | MAY_WRITE | MAY_EXEC;
 	if (mask == MAY_READ)
-		if (inode_capable(inode, CAP_DAC_READ_SEARCH))
+		if (capable_wrt_inode_uidgid(inode, CAP_DAC_READ_SEARCH))
 			return 0;
 
 	return -EACCES;
@@ -2199,7 +2200,7 @@ static inline int check_sticky(struct inode *dir, struct inode *inode)
 		return 0;
 	if (uid_eq(dir->i_uid, fsuid))
 		return 0;
-	return !inode_capable(inode, CAP_FOWNER);
+	return !capable_wrt_inode_uidgid(inode, CAP_FOWNER);
 }
 
 /*

include/linux/capability.h

@@ -211,7 +211,7 @@ extern bool has_ns_capability_noaudit(struct task_struct *t,
 extern bool capable(int cap);
 extern bool ns_capable(struct user_namespace *ns, int cap);
 extern bool nsown_capable(int cap);
-extern bool inode_capable(const struct inode *inode, int cap);
+extern bool capable_wrt_inode_uidgid(const struct inode *inode, int cap);
 extern bool file_ns_capable(const struct file *file, struct user_namespace *ns, int cap);
 
 /* audit system wants to get cap info from files as well */

kernel/auditsc.c

@@ -733,6 +733,22 @@ static enum audit_state audit_filter_task(struct task_struct *tsk, char **key)
 	return AUDIT_BUILD_CONTEXT;
 }
 
+static int audit_in_mask(const struct audit_krule *rule, unsigned long val)
+{
+	int word, bit;
+
+	if (val > 0xffffffff)
+		return false;
+
+	word = AUDIT_WORD(val);
+	if (word >= AUDIT_BITMASK_SIZE)
+		return false;
+
+	bit = AUDIT_BIT(val);
+
+	return rule->mask[word] & bit;
+}
+
 /* At syscall entry and exit time, this filter is called if the
  * audit_state is not low enough that auditing cannot take place, but is
  * also not high enough that we already know we have to write an audit
@@ -750,11 +766,8 @@ static enum audit_state audit_filter_syscall(struct task_struct *tsk,
 
 	rcu_read_lock();
 	if (!list_empty(list)) {
-		int word = AUDIT_WORD(ctx->major);
-		int bit  = AUDIT_BIT(ctx->major);
-
 		list_for_each_entry_rcu(e, list, list) {
-			if ((e->rule.mask[word] & bit) == bit &&
+			if (audit_in_mask(&e->rule, ctx->major) &&
 			    audit_filter_rules(tsk, &e->rule, ctx, NULL,
 					       &state, false)) {
 				rcu_read_unlock();
@@ -774,20 +787,16 @@ static enum audit_state audit_filter_syscall(struct task_struct *tsk,
 static int audit_filter_inode_name(struct task_struct *tsk,
 				   struct audit_names *n,
 				   struct audit_context *ctx) {
-	int word, bit;
 	int h = audit_hash_ino((u32)n->ino);
 	struct list_head *list = &audit_inode_hash[h];
 	struct audit_entry *e;
 	enum audit_state state;
 
-	word = AUDIT_WORD(ctx->major);
-	bit  = AUDIT_BIT(ctx->major);
-
 	if (list_empty(list))
 		return 0;
 
 	list_for_each_entry_rcu(e, list, list) {
-		if ((e->rule.mask[word] & bit) == bit &&
+		if (audit_in_mask(&e->rule, ctx->major) &&
 		    audit_filter_rules(tsk, &e->rule, ctx, n, &state, false)) {
 			ctx->current_state = state;
 			return 1;

kernel/capability.c

@@ -445,22 +445,18 @@ bool nsown_capable(int cap)
 }
 
 /**
- * inode_capable - Check superior capability over inode
+ * capable_wrt_inode_uidgid - Check nsown_capable and uid and gid mapped
  * @inode: The inode in question
  * @cap: The capability in question
  *
- * Return true if the current task has the given superior capability
- * targeted at it's own user namespace and that the given inode is owned
- * by the current user namespace or a child namespace.
- *
- * Currently we check to see if an inode is owned by the current
- * user namespace by seeing if the inode's owner maps into the
- * current user namespace.
- *
+ * Return true if the current task has the given capability targeted at
+ * its own user namespace and that the given inode's uid and gid are
+ * mapped into the current user namespace.
  */
-bool inode_capable(const struct inode *inode, int cap)
+bool capable_wrt_inode_uidgid(const struct inode *inode, int cap)
 {
 	struct user_namespace *ns = current_user_ns();
 
-	return ns_capable(ns, cap) && kuid_has_mapping(ns, inode->i_uid);
+	return ns_capable(ns, cap) && kuid_has_mapping(ns, inode->i_uid) &&
+		kgid_has_mapping(ns, inode->i_gid);
 }

mm/compaction.c

@@ -657,17 +657,21 @@ static void isolate_freepages(struct zone *zone,
 				struct compact_control *cc)
 {
 	struct page *page;
-	unsigned long high_pfn, low_pfn, pfn, z_end_pfn, end_pfn;
+	unsigned long high_pfn, low_pfn, pfn, z_end_pfn;
 	int nr_freepages = cc->nr_freepages;
 	struct list_head *freelist = &cc->freepages;
 
 	/*
 	 * Initialise the free scanner. The starting point is where we last
-	 * scanned from (or the end of the zone if starting). The low point
-	 * is the end of the pageblock the migration scanner is using.
+	 * successfully isolated from, zone-cached value, or the end of the
+	 * zone when isolating for the first time. We need this aligned to
+	 * the pageblock boundary, because we do pfn -= pageblock_nr_pages
+	 * in the for loop.
+	 * The low boundary is the end of the pageblock the migration scanner
+	 * is using.
 	 */
-	pfn = cc->free_pfn;
-	low_pfn = cc->migrate_pfn + pageblock_nr_pages;
+	pfn = cc->free_pfn & ~(pageblock_nr_pages-1);
+	low_pfn = ALIGN(cc->migrate_pfn + 1, pageblock_nr_pages);
 
 	/*
 	 * Take care that if the migration scanner is at the end of the zone
@@ -683,9 +687,10 @@ static void isolate_freepages(struct zone *zone,
 	 * pages on cc->migratepages. We stop searching if the migrate
 	 * and free page scanners meet or enough free pages are isolated.
 	 */
-	for (; pfn > low_pfn && cc->nr_migratepages > nr_freepages;
+	for (; pfn >= low_pfn && cc->nr_migratepages > nr_freepages;
 					pfn -= pageblock_nr_pages) {
 		unsigned long isolated;
+		unsigned long end_pfn;
 
 		if (!pfn_valid(pfn))
 			continue;
@@ -713,13 +718,10 @@ static void isolate_freepages(struct zone *zone,
 		isolated = 0;
 
 		/*
-		 * As pfn may not start aligned, pfn+pageblock_nr_page
-		 * may cross a MAX_ORDER_NR_PAGES boundary and miss
-		 * a pfn_valid check. Ensure isolate_freepages_block()
-		 * only scans within a pageblock
+		 * Take care when isolating in last pageblock of a zone which
+		 * ends in the middle of a pageblock.
 		 */
-		end_pfn = ALIGN(pfn + 1, pageblock_nr_pages);
-		end_pfn = min(end_pfn, z_end_pfn);
+		end_pfn = min(pfn + pageblock_nr_pages, z_end_pfn);
 		isolated = isolate_freepages_block(cc, pfn, end_pfn,
 						   freelist, false);
 		nr_freepages += isolated;
@@ -738,7 +740,14 @@ static void isolate_freepages(struct zone *zone,
 	/* split_free_page does not map the pages */
 	map_pages(freelist);
 
-	cc->free_pfn = high_pfn;
+	/*
+	 * If we crossed the migrate scanner, we want to keep it that way
+	 * so that compact_finished() may detect this
+	 */
+	if (pfn < low_pfn)
+		cc->free_pfn = max(pfn, zone->zone_start_pfn);
+	else
+		cc->free_pfn = high_pfn;
 	cc->nr_freepages = nr_freepages;
 }
 
@@ -946,6 +955,14 @@ static int compact_zone(struct zone *zone, struct compact_control *cc)
 		;
 	}
 
+	/*
+	 * Clear pageblock skip if there were failures recently and compaction
+	 * is about to be retried after being deferred. kswapd does not do
+	 * this reset as it'll reset the cached information when going to sleep.
+	 */
+	if (compaction_restarting(zone, cc->order) && !current_is_kswapd())
+		__reset_isolation_suitable(zone);
+
 	/*
 	 * Setup to move all movable pages to the end of the zone. Used cached
 	 * information on where the scanners should start but check that it
@@ -962,14 +979,6 @@ static int compact_zone(struct zone *zone, struct compact_control *cc)
 		zone->compact_cached_migrate_pfn = cc->migrate_pfn;
 	}
 
-	/*
-	 * Clear pageblock skip if there were failures recently and compaction
-	 * is about to be retried after being deferred. kswapd does not do
-	 * this reset as it'll reset the cached information when going to sleep.
-	 */
-	if (compaction_restarting(zone, cc->order) && !current_is_kswapd())
-		__reset_isolation_suitable(zone);
-
 	migrate_prep_local();
 
 	while ((ret = compact_finished(zone, cc)) == COMPACT_CONTINUE) {
@@ -1003,7 +1012,11 @@ static int compact_zone(struct zone *zone, struct compact_control *cc)
 		if (err) {
 			putback_movable_pages(&cc->migratepages);
 			cc->nr_migratepages = 0;
-			if (err == -ENOMEM) {
+			/*
+			 * migrate_pages() may return -ENOMEM when scanners meet
+			 * and we want compact_finished() to detect it
+			 */
+			if (err == -ENOMEM && cc->free_pfn > cc->migrate_pfn) {
 				ret = COMPACT_PARTIAL;
 				goto out;
 			}

net/ipv4/netfilter/nf_defrag_ipv4.c

@@ -22,7 +22,6 @@
 #endif
 #include <net/netfilter/nf_conntrack_zones.h>
 
-/* Returns new sk_buff, or NULL */
 static int nf_ct_ipv4_gather_frags(struct sk_buff *skb, u_int32_t user)
 {
 	int err;
@@ -33,8 +32,10 @@ static int nf_ct_ipv4_gather_frags(struct sk_buff *skb, u_int32_t user)
 	err = ip_defrag(skb, user);
 	local_bh_enable();
 
-	if (!err)
+	if (!err) {
 		ip_send_check(ip_hdr(skb));
+		skb->local_df = 1;
+	}
 
 	return err;
 }