Merge branch 'for-linus' of master.kernel.org:/pub/scm/linux/kernel/git/roland/infiniband

drivers/infiniband/core/cm.c

@@ -3163,22 +3163,6 @@ int ib_cm_init_qp_attr(struct ib_cm_id *cm_id,
 }
 EXPORT_SYMBOL(ib_cm_init_qp_attr);
 
-static __be64 cm_get_ca_guid(struct ib_device *device)
-{
-	struct ib_device_attr *device_attr;
-	__be64 guid;
-	int ret;
-
-	device_attr = kmalloc(sizeof *device_attr, GFP_KERNEL);
-	if (!device_attr)
-		return 0;
-
-	ret = ib_query_device(device, device_attr);
-	guid = ret ? 0 : device_attr->node_guid;
-	kfree(device_attr);
-	return guid;
-}
-
 static void cm_add_one(struct ib_device *device)
 {
 	struct cm_device *cm_dev;
@@ -3200,9 +3184,7 @@ static void cm_add_one(struct ib_device *device)
 		return;
 
 	cm_dev->device = device;
-	cm_dev->ca_guid = cm_get_ca_guid(device);
-	if (!cm_dev->ca_guid)
-		goto error1;
+	cm_dev->ca_guid = device->node_guid;
 
 	set_bit(IB_MGMT_METHOD_SEND, reg_req.method_mask);
 	for (i = 1; i <= device->phys_port_cnt; i++) {
@@ -3217,11 +3199,11 @@ static void cm_add_one(struct ib_device *device)
 							cm_recv_handler,
 							port);
 		if (IS_ERR(port->mad_agent))
-			goto error2;
+			goto error1;
 
 		ret = ib_modify_port(device, i, 0, &port_modify);
 		if (ret)
-			goto error3;
+			goto error2;
 	}
 	ib_set_client_data(device, &cm_client, cm_dev);
 
@@ -3230,9 +3212,9 @@ static void cm_add_one(struct ib_device *device)
 	write_unlock_irqrestore(&cm.device_lock, flags);
 	return;
 
-error3:
-	ib_unregister_mad_agent(port->mad_agent);
 error2:
+	ib_unregister_mad_agent(port->mad_agent);
+error1:
 	port_modify.set_port_cap_mask = 0;
 	port_modify.clr_port_cap_mask = IB_PORT_CM_SUP;
 	while (--i) {
@@ -3240,7 +3222,6 @@ static void cm_add_one(struct ib_device *device)
 		ib_modify_port(device, port->port_num, 0, &port_modify);
 		ib_unregister_mad_agent(port->mad_agent);
 	}
-error1:
 	kfree(cm_dev);
 }
 

drivers/infiniband/core/device.c

@@ -38,8 +38,7 @@
 #include <linux/errno.h>
 #include <linux/slab.h>
 #include <linux/init.h>
-
-#include <asm/semaphore.h>
+#include <linux/mutex.h>
 
 #include "core_priv.h"
 
@@ -57,13 +56,13 @@ static LIST_HEAD(device_list);
 static LIST_HEAD(client_list);
 
 /*
- * device_sem protects access to both device_list and client_list.
+ * device_mutex protects access to both device_list and client_list.
  * There's no real point to using multiple locks or something fancier
  * like an rwsem: we always access both lists, and we're always
  * modifying one list or the other list.  In any case this is not a
  * hot path so there's no point in trying to optimize.
  */
-static DECLARE_MUTEX(device_sem);
+static DEFINE_MUTEX(device_mutex);
 
 static int ib_device_check_mandatory(struct ib_device *device)
 {
@@ -221,7 +220,7 @@ int ib_register_device(struct ib_device *device)
 {
 	int ret;
 
-	down(&device_sem);
+	mutex_lock(&device_mutex);
 
 	if (strchr(device->name, '%')) {
 		ret = alloc_name(device->name);
@@ -259,7 +258,7 @@ int ib_register_device(struct ib_device *device)
 	}
 
  out:
-	up(&device_sem);
+	mutex_unlock(&device_mutex);
 	return ret;
 }
 EXPORT_SYMBOL(ib_register_device);
@@ -276,7 +275,7 @@ void ib_unregister_device(struct ib_device *device)
 	struct ib_client_data *context, *tmp;
 	unsigned long flags;
 
-	down(&device_sem);
+	mutex_lock(&device_mutex);
 
 	list_for_each_entry_reverse(client, &client_list, list)
 		if (client->remove)
@@ -284,7 +283,7 @@ void ib_unregister_device(struct ib_device *device)
 
 	list_del(&device->core_list);
 
-	up(&device_sem);
+	mutex_unlock(&device_mutex);
 
 	spin_lock_irqsave(&device->client_data_lock, flags);
 	list_for_each_entry_safe(context, tmp, &device->client_data_list, list)
@@ -312,14 +311,14 @@ int ib_register_client(struct ib_client *client)
 {
 	struct ib_device *device;
 
-	down(&device_sem);
+	mutex_lock(&device_mutex);
 
 	list_add_tail(&client->list, &client_list);
 	list_for_each_entry(device, &device_list, core_list)
 		if (client->add && !add_client_context(device, client))
 			client->add(device);
 
-	up(&device_sem);
+	mutex_unlock(&device_mutex);
 
 	return 0;
 }
@@ -339,7 +338,7 @@ void ib_unregister_client(struct ib_client *client)
 	struct ib_device *device;
 	unsigned long flags;
 
-	down(&device_sem);
+	mutex_lock(&device_mutex);
 
 	list_for_each_entry(device, &device_list, core_list) {
 		if (client->remove)
@@ -355,7 +354,7 @@ void ib_unregister_client(struct ib_client *client)
 	}
 	list_del(&client->list);
 
-	up(&device_sem);
+	mutex_unlock(&device_mutex);
 }
 EXPORT_SYMBOL(ib_unregister_client);
 

drivers/infiniband/core/sysfs.c

@@ -445,13 +445,7 @@ static int ib_device_uevent(struct class_device *cdev, char **envp,
 		return -ENOMEM;
 
 	/*
-	 * It might be nice to pass the node GUID with the event, but
-	 * right now the only way to get it is to query the device
-	 * provider, and this can crash during device removal because
-	 * we are will be running after driver removal has started.
-	 * We could add a node_guid field to struct ib_device, or we
-	 * could just let userspace read the node GUID from sysfs when
-	 * devices are added.
+	 * It would be nice to pass the node GUID with the event...
 	 */
 
 	envp[i] = NULL;
@@ -623,21 +617,15 @@ static ssize_t show_sys_image_guid(struct class_device *cdev, char *buf)
 static ssize_t show_node_guid(struct class_device *cdev, char *buf)
 {
 	struct ib_device *dev = container_of(cdev, struct ib_device, class_dev);
-	struct ib_device_attr attr;
-	ssize_t ret;
 
 	if (!ibdev_is_alive(dev))
 		return -ENODEV;
 
-	ret = ib_query_device(dev, &attr);
-	if (ret)
-		return ret;
-
 	return sprintf(buf, "%04x:%04x:%04x:%04x\n",
-		       be16_to_cpu(((__be16 *) &attr.node_guid)[0]),
-		       be16_to_cpu(((__be16 *) &attr.node_guid)[1]),
-		       be16_to_cpu(((__be16 *) &attr.node_guid)[2]),
-		       be16_to_cpu(((__be16 *) &attr.node_guid)[3]));
+		       be16_to_cpu(((__be16 *) &dev->node_guid)[0]),
+		       be16_to_cpu(((__be16 *) &dev->node_guid)[1]),
+		       be16_to_cpu(((__be16 *) &dev->node_guid)[2]),
+		       be16_to_cpu(((__be16 *) &dev->node_guid)[3]));
 }
 
 static CLASS_DEVICE_ATTR(node_type, S_IRUGO, show_node_type, NULL);

drivers/infiniband/core/ucm.c

@@ -42,6 +42,7 @@
 #include <linux/mount.h>
 #include <linux/cdev.h>
 #include <linux/idr.h>
+#include <linux/mutex.h>
 
 #include <asm/uaccess.h>
 
@@ -113,7 +114,7 @@ static struct ib_client ucm_client = {
 	.remove = ib_ucm_remove_one
 };
 
-static DECLARE_MUTEX(ctx_id_mutex);
+static DEFINE_MUTEX(ctx_id_mutex);
 static DEFINE_IDR(ctx_id_table);
 static DECLARE_BITMAP(dev_map, IB_UCM_MAX_DEVICES);
 
@@ -121,7 +122,7 @@ static struct ib_ucm_context *ib_ucm_ctx_get(struct ib_ucm_file *file, int id)
 {
 	struct ib_ucm_context *ctx;
 
-	down(&ctx_id_mutex);
+	mutex_lock(&ctx_id_mutex);
 	ctx = idr_find(&ctx_id_table, id);
 	if (!ctx)
 		ctx = ERR_PTR(-ENOENT);
@@ -129,7 +130,7 @@ static struct ib_ucm_context *ib_ucm_ctx_get(struct ib_ucm_file *file, int id)
 		ctx = ERR_PTR(-EINVAL);
 	else
 		atomic_inc(&ctx->ref);
-	up(&ctx_id_mutex);
+	mutex_unlock(&ctx_id_mutex);
 
 	return ctx;
 }
@@ -186,9 +187,9 @@ static struct ib_ucm_context *ib_ucm_ctx_alloc(struct ib_ucm_file *file)
 		if (!result)
 			goto error;
 
-		down(&ctx_id_mutex);
+		mutex_lock(&ctx_id_mutex);
 		result = idr_get_new(&ctx_id_table, ctx, &ctx->id);
-		up(&ctx_id_mutex);
+		mutex_unlock(&ctx_id_mutex);
 	} while (result == -EAGAIN);
 
 	if (result)
@@ -550,9 +551,9 @@ static ssize_t ib_ucm_create_id(struct ib_ucm_file *file,
 err2:
 	ib_destroy_cm_id(ctx->cm_id);
 err1:
-	down(&ctx_id_mutex);
+	mutex_lock(&ctx_id_mutex);
 	idr_remove(&ctx_id_table, ctx->id);
-	up(&ctx_id_mutex);
+	mutex_unlock(&ctx_id_mutex);
 	kfree(ctx);
 	return result;
 }
@@ -572,7 +573,7 @@ static ssize_t ib_ucm_destroy_id(struct ib_ucm_file *file,
 	if (copy_from_user(&cmd, inbuf, sizeof(cmd)))
 		return -EFAULT;
 
-	down(&ctx_id_mutex);
+	mutex_lock(&ctx_id_mutex);
 	ctx = idr_find(&ctx_id_table, cmd.id);
 	if (!ctx)
 		ctx = ERR_PTR(-ENOENT);
@@ -580,7 +581,7 @@ static ssize_t ib_ucm_destroy_id(struct ib_ucm_file *file,
 		ctx = ERR_PTR(-EINVAL);
 	else
 		idr_remove(&ctx_id_table, ctx->id);
-	up(&ctx_id_mutex);
+	mutex_unlock(&ctx_id_mutex);
 
 	if (IS_ERR(ctx))
 		return PTR_ERR(ctx);
@@ -1280,9 +1281,9 @@ static int ib_ucm_close(struct inode *inode, struct file *filp)
 				 struct ib_ucm_context, file_list);
 		up(&file->mutex);
 
-		down(&ctx_id_mutex);
+		mutex_lock(&ctx_id_mutex);
 		idr_remove(&ctx_id_table, ctx->id);
-		up(&ctx_id_mutex);
+		mutex_unlock(&ctx_id_mutex);
 
 		ib_destroy_cm_id(ctx->cm_id);
 		ib_ucm_cleanup_events(ctx);

drivers/infiniband/core/uverbs.h

@@ -41,6 +41,7 @@
 
 #include <linux/kref.h>
 #include <linux/idr.h>
+#include <linux/mutex.h>
 
 #include <rdma/ib_verbs.h>
 #include <rdma/ib_user_verbs.h>
@@ -88,7 +89,7 @@ struct ib_uverbs_event_file {
 
 struct ib_uverbs_file {
 	struct kref				ref;
-	struct semaphore			mutex;
+	struct mutex				mutex;
 	struct ib_uverbs_device		       *device;
 	struct ib_ucontext		       *ucontext;
 	struct ib_event_handler			event_handler;
@@ -131,7 +132,7 @@ struct ib_ucq_object {
 	u32			async_events_reported;
 };
 
-extern struct semaphore ib_uverbs_idr_mutex;
+extern struct mutex ib_uverbs_idr_mutex;
 extern struct idr ib_uverbs_pd_idr;
 extern struct idr ib_uverbs_mr_idr;
 extern struct idr ib_uverbs_mw_idr;

drivers/infiniband/core/uverbs_main.c

@@ -66,7 +66,7 @@ enum {
 
 static struct class *uverbs_class;
 
-DECLARE_MUTEX(ib_uverbs_idr_mutex);
+DEFINE_MUTEX(ib_uverbs_idr_mutex);
 DEFINE_IDR(ib_uverbs_pd_idr);
 DEFINE_IDR(ib_uverbs_mr_idr);
 DEFINE_IDR(ib_uverbs_mw_idr);
@@ -180,7 +180,7 @@ static int ib_uverbs_cleanup_ucontext(struct ib_uverbs_file *file,
 	if (!context)
 		return 0;
 
-	down(&ib_uverbs_idr_mutex);
+	mutex_lock(&ib_uverbs_idr_mutex);
 
 	list_for_each_entry_safe(uobj, tmp, &context->ah_list, list) {
 		struct ib_ah *ah = idr_find(&ib_uverbs_ah_idr, uobj->id);
@@ -250,7 +250,7 @@ static int ib_uverbs_cleanup_ucontext(struct ib_uverbs_file *file,
 		kfree(uobj);
 	}
 
-	up(&ib_uverbs_idr_mutex);
+	mutex_unlock(&ib_uverbs_idr_mutex);
 
 	return context->device->dealloc_ucontext(context);
 }
@@ -653,7 +653,7 @@ static int ib_uverbs_open(struct inode *inode, struct file *filp)
 	file->ucontext	 = NULL;
 	file->async_file = NULL;
 	kref_init(&file->ref);
-	init_MUTEX(&file->mutex);
+	mutex_init(&file->mutex);
 
 	filp->private_data = file;
 

drivers/infiniband/hw/mthca/mthca_av.c

@@ -163,6 +163,11 @@ int mthca_destroy_ah(struct mthca_dev *dev, struct mthca_ah *ah)
 	return 0;
 }
 
+int mthca_ah_grh_present(struct mthca_ah *ah)
+{
+	return !!(ah->av->g_slid & 0x80);
+}
+
 int mthca_read_ah(struct mthca_dev *dev, struct mthca_ah *ah,
 		  struct ib_ud_header *header)
 {
@@ -172,8 +177,7 @@ int mthca_read_ah(struct mthca_dev *dev, struct mthca_ah *ah,
 	header->lrh.service_level   = be32_to_cpu(ah->av->sl_tclass_flowlabel) >> 28;
 	header->lrh.destination_lid = ah->av->dlid;
 	header->lrh.source_lid      = cpu_to_be16(ah->av->g_slid & 0x7f);
-	if (ah->av->g_slid & 0x80) {
-		header->grh_present = 1;
+	if (mthca_ah_grh_present(ah)) {
 		header->grh.traffic_class =
 			(be32_to_cpu(ah->av->sl_tclass_flowlabel) >> 20) & 0xff;
 		header->grh.flow_label    =
@@ -184,8 +188,6 @@ int mthca_read_ah(struct mthca_dev *dev, struct mthca_ah *ah,
 				  &header->grh.source_gid);
 		memcpy(header->grh.destination_gid.raw,
 		       ah->av->dgid, 16);
-	} else {
-		header->grh_present = 0;
 	}
 
 	return 0;

drivers/infiniband/hw/mthca/mthca_cmd.c

@@ -606,7 +606,7 @@ static int mthca_map_cmd(struct mthca_dev *dev, u16 op, struct mthca_icm *icm,
 			err = -EINVAL;
 			goto out;
 		}
-		for (i = 0; i < mthca_icm_size(&iter) / (1 << lg); ++i) {
+		for (i = 0; i < mthca_icm_size(&iter) >> lg; ++i) {
 			if (virt != -1) {
 				pages[nent * 2] = cpu_to_be64(virt);
 				virt += 1 << lg;
@@ -727,8 +727,8 @@ int mthca_QUERY_FW(struct mthca_dev *dev, u8 *status)
 		 * system pages needed.
 		 */
 		dev->fw.arbel.fw_pages =
-			(dev->fw.arbel.fw_pages + (1 << (PAGE_SHIFT - 12)) - 1) >>
-			(PAGE_SHIFT - 12);
+			ALIGN(dev->fw.arbel.fw_pages, PAGE_SIZE >> 12) >>
+				(PAGE_SHIFT - 12);
 
 		mthca_dbg(dev, "Clear int @ %llx, EQ arm @ %llx, EQ set CI @ %llx\n",
 			  (unsigned long long) dev->fw.arbel.clr_int_base,
@@ -1445,6 +1445,7 @@ int mthca_SET_ICM_SIZE(struct mthca_dev *dev, u64 icm_size, u64 *aux_pages,
 	 * pages needed.
 	 */
 	*aux_pages = (*aux_pages + (1 << (PAGE_SHIFT - 12)) - 1) >> (PAGE_SHIFT - 12);
+	*aux_pages = ALIGN(*aux_pages, PAGE_SIZE >> 12) >> (PAGE_SHIFT - 12);
 
 	return 0;
 }

drivers/infiniband/hw/mthca/mthca_dev.h

@@ -520,6 +520,7 @@ int mthca_create_ah(struct mthca_dev *dev,
 int mthca_destroy_ah(struct mthca_dev *dev, struct mthca_ah *ah);
 int mthca_read_ah(struct mthca_dev *dev, struct mthca_ah *ah,
 		  struct ib_ud_header *header);
+int mthca_ah_grh_present(struct mthca_ah *ah);
 
 int mthca_multicast_attach(struct ib_qp *ibqp, union ib_gid *gid, u16 lid);
 int mthca_multicast_detach(struct ib_qp *ibqp, union ib_gid *gid, u16 lid);

drivers/infiniband/hw/mthca/mthca_eq.c

@@ -45,6 +45,7 @@
 enum {
 	MTHCA_NUM_ASYNC_EQE = 0x80,
 	MTHCA_NUM_CMD_EQE   = 0x80,
+	MTHCA_NUM_SPARE_EQE = 0x80,
 	MTHCA_EQ_ENTRY_SIZE = 0x20
 };
 
@@ -277,11 +278,10 @@ static int mthca_eq_int(struct mthca_dev *dev, struct mthca_eq *eq)
 {
 	struct mthca_eqe *eqe;
 	int disarm_cqn;
-	int  eqes_found = 0;
+	int eqes_found = 0;
+	int set_ci = 0;
 
 	while ((eqe = next_eqe_sw(eq))) {
-		int set_ci = 0;
-
 		/*
 		 * Make sure we read EQ entry contents after we've
 		 * checked the ownership bit.
@@ -345,12 +345,6 @@ static int mthca_eq_int(struct mthca_dev *dev, struct mthca_eq *eq)
 					be16_to_cpu(eqe->event.cmd.token),
 					eqe->event.cmd.status,
 					be64_to_cpu(eqe->event.cmd.out_param));
-			/*
-			 * cmd_event() may add more commands.
-			 * The card will think the queue has overflowed if
-			 * we don't tell it we've been processing events.
-			 */
-			set_ci = 1;
 			break;
 
 		case MTHCA_EVENT_TYPE_PORT_CHANGE:
@@ -385,8 +379,16 @@ static int mthca_eq_int(struct mthca_dev *dev, struct mthca_eq *eq)
 		set_eqe_hw(eqe);
 		++eq->cons_index;
 		eqes_found = 1;
+		++set_ci;
 
-		if (unlikely(set_ci)) {
+		/*
+		 * The HCA will think the queue has overflowed if we
+		 * don't tell it we've been processing events.  We
+		 * create our EQs with MTHCA_NUM_SPARE_EQE extra
+		 * entries, so we must update our consumer index at
+		 * least that often.
+		 */
+		if (unlikely(set_ci >= MTHCA_NUM_SPARE_EQE)) {
 			/*
 			 * Conditional on hca_type is OK here because
 			 * this is a rare case, not the fast path.
@@ -862,19 +864,19 @@ int __devinit mthca_init_eq_table(struct mthca_dev *dev)
 	intr = (dev->mthca_flags & MTHCA_FLAG_MSI) ?
 		128 : dev->eq_table.inta_pin;
 
-	err = mthca_create_eq(dev, dev->limits.num_cqs,
+	err = mthca_create_eq(dev, dev->limits.num_cqs + MTHCA_NUM_SPARE_EQE,
 			      (dev->mthca_flags & MTHCA_FLAG_MSI_X) ? 128 : intr,
 			      &dev->eq_table.eq[MTHCA_EQ_COMP]);
 	if (err)
 		goto err_out_unmap;
 
-	err = mthca_create_eq(dev, MTHCA_NUM_ASYNC_EQE,
+	err = mthca_create_eq(dev, MTHCA_NUM_ASYNC_EQE + MTHCA_NUM_SPARE_EQE,
 			      (dev->mthca_flags & MTHCA_FLAG_MSI_X) ? 129 : intr,
 			      &dev->eq_table.eq[MTHCA_EQ_ASYNC]);
 	if (err)
 		goto err_out_comp;
 
-	err = mthca_create_eq(dev, MTHCA_NUM_CMD_EQE,
+	err = mthca_create_eq(dev, MTHCA_NUM_CMD_EQE + MTHCA_NUM_SPARE_EQE,
 			      (dev->mthca_flags & MTHCA_FLAG_MSI_X) ? 130 : intr,
 			      &dev->eq_table.eq[MTHCA_EQ_CMD]);
 	if (err)

drivers/infiniband/hw/mthca/mthca_provider.c

@@ -33,7 +33,7 @@
  * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
  * SOFTWARE.
  *
- * $Id: mthca_provider.c 1397 2004-12-28 05:09:00Z roland $
+ * $Id: mthca_provider.c 4859 2006-01-09 21:55:10Z roland $
  */
 
 #include <rdma/ib_smi.h>
@@ -45,6 +45,14 @@
 #include "mthca_user.h"
 #include "mthca_memfree.h"
 
+static void init_query_mad(struct ib_smp *mad)
+{
+	mad->base_version  = 1;
+	mad->mgmt_class    = IB_MGMT_CLASS_SUBN_LID_ROUTED;
+	mad->class_version = 1;
+	mad->method    	   = IB_MGMT_METHOD_GET;
+}
+
 static int mthca_query_device(struct ib_device *ibdev,
 			      struct ib_device_attr *props)
 {
@@ -55,7 +63,7 @@ static int mthca_query_device(struct ib_device *ibdev,
 
 	u8 status;
 
-	in_mad  = kmalloc(sizeof *in_mad, GFP_KERNEL);
+	in_mad  = kzalloc(sizeof *in_mad, GFP_KERNEL);
 	out_mad = kmalloc(sizeof *out_mad, GFP_KERNEL);
 	if (!in_mad || !out_mad)
 		goto out;
@@ -64,12 +72,8 @@ static int mthca_query_device(struct ib_device *ibdev,
 
 	props->fw_ver              = mdev->fw_ver;
 
-	memset(in_mad, 0, sizeof *in_mad);
-	in_mad->base_version       = 1;
-	in_mad->mgmt_class     	   = IB_MGMT_CLASS_SUBN_LID_ROUTED;
-	in_mad->class_version  	   = 1;
-	in_mad->method         	   = IB_MGMT_METHOD_GET;
-	in_mad->attr_id   	   = IB_SMP_ATTR_NODE_INFO;
+	init_query_mad(in_mad);
+	in_mad->attr_id = IB_SMP_ATTR_NODE_INFO;
 
 	err = mthca_MAD_IFC(mdev, 1, 1,
 			    1, NULL, NULL, in_mad, out_mad,
@@ -87,7 +91,6 @@ static int mthca_query_device(struct ib_device *ibdev,
 	props->vendor_part_id      = be16_to_cpup((__be16 *) (out_mad->data + 30));
 	props->hw_ver              = be32_to_cpup((__be32 *) (out_mad->data + 32));
 	memcpy(&props->sys_image_guid, out_mad->data +  4, 8);
-	memcpy(&props->node_guid,      out_mad->data + 12, 8);
 
 	props->max_mr_size         = ~0ull;
 	props->page_size_cap       = mdev->limits.page_size_cap;
@@ -128,20 +131,16 @@ static int mthca_query_port(struct ib_device *ibdev,
 	int err = -ENOMEM;
 	u8 status;
 
-	in_mad  = kmalloc(sizeof *in_mad, GFP_KERNEL);
+	in_mad  = kzalloc(sizeof *in_mad, GFP_KERNEL);
 	out_mad = kmalloc(sizeof *out_mad, GFP_KERNEL);
 	if (!in_mad || !out_mad)
 		goto out;
 
 	memset(props, 0, sizeof *props);
 
-	memset(in_mad, 0, sizeof *in_mad);
-	in_mad->base_version       = 1;
-	in_mad->mgmt_class     	   = IB_MGMT_CLASS_SUBN_LID_ROUTED;
-	in_mad->class_version  	   = 1;
-	in_mad->method         	   = IB_MGMT_METHOD_GET;
-	in_mad->attr_id   	   = IB_SMP_ATTR_PORT_INFO;
-	in_mad->attr_mod           = cpu_to_be32(port);
+	init_query_mad(in_mad);
+	in_mad->attr_id  = IB_SMP_ATTR_PORT_INFO;
+	in_mad->attr_mod = cpu_to_be32(port);
 
 	err = mthca_MAD_IFC(to_mdev(ibdev), 1, 1,
 			    port, NULL, NULL, in_mad, out_mad,
@@ -220,18 +219,14 @@ static int mthca_query_pkey(struct ib_device *ibdev,
 	int err = -ENOMEM;
 	u8 status;
 
-	in_mad  = kmalloc(sizeof *in_mad, GFP_KERNEL);
+	in_mad  = kzalloc(sizeof *in_mad, GFP_KERNEL);
 	out_mad = kmalloc(sizeof *out_mad, GFP_KERNEL);
 	if (!in_mad || !out_mad)
 		goto out;
 
-	memset(in_mad, 0, sizeof *in_mad);
-	in_mad->base_version       = 1;
-	in_mad->mgmt_class     	   = IB_MGMT_CLASS_SUBN_LID_ROUTED;
-	in_mad->class_version  	   = 1;
-	in_mad->method         	   = IB_MGMT_METHOD_GET;
-	in_mad->attr_id   	   = IB_SMP_ATTR_PKEY_TABLE;
-	in_mad->attr_mod           = cpu_to_be32(index / 32);
+	init_query_mad(in_mad);
+	in_mad->attr_id  = IB_SMP_ATTR_PKEY_TABLE;
+	in_mad->attr_mod = cpu_to_be32(index / 32);
 
 	err = mthca_MAD_IFC(to_mdev(ibdev), 1, 1,
 			    port, NULL, NULL, in_mad, out_mad,
@@ -259,18 +254,14 @@ static int mthca_query_gid(struct ib_device *ibdev, u8 port,
 	int err = -ENOMEM;
 	u8 status;
 
-	in_mad  = kmalloc(sizeof *in_mad, GFP_KERNEL);
+	in_mad  = kzalloc(sizeof *in_mad, GFP_KERNEL);
 	out_mad = kmalloc(sizeof *out_mad, GFP_KERNEL);
 	if (!in_mad || !out_mad)
 		goto out;
 
-	memset(in_mad, 0, sizeof *in_mad);
-	in_mad->base_version       = 1;
-	in_mad->mgmt_class     	   = IB_MGMT_CLASS_SUBN_LID_ROUTED;
-	in_mad->class_version  	   = 1;
-	in_mad->method         	   = IB_MGMT_METHOD_GET;
-	in_mad->attr_id   	   = IB_SMP_ATTR_PORT_INFO;
-	in_mad->attr_mod           = cpu_to_be32(port);
+	init_query_mad(in_mad);
+	in_mad->attr_id  = IB_SMP_ATTR_PORT_INFO;
+	in_mad->attr_mod = cpu_to_be32(port);
 
 	err = mthca_MAD_IFC(to_mdev(ibdev), 1, 1,
 			    port, NULL, NULL, in_mad, out_mad,
@@ -284,13 +275,9 @@ static int mthca_query_gid(struct ib_device *ibdev, u8 port,
 
 	memcpy(gid->raw, out_mad->data + 8, 8);
 
-	memset(in_mad, 0, sizeof *in_mad);
-	in_mad->base_version       = 1;
-	in_mad->mgmt_class     	   = IB_MGMT_CLASS_SUBN_LID_ROUTED;
-	in_mad->class_version  	   = 1;
-	in_mad->method         	   = IB_MGMT_METHOD_GET;
-	in_mad->attr_id   	   = IB_SMP_ATTR_GUID_INFO;
-	in_mad->attr_mod           = cpu_to_be32(index / 8);
+	init_query_mad(in_mad);
+	in_mad->attr_id  = IB_SMP_ATTR_GUID_INFO;
+	in_mad->attr_mod = cpu_to_be32(index / 8);
 
 	err = mthca_MAD_IFC(to_mdev(ibdev), 1, 1,
 			    port, NULL, NULL, in_mad, out_mad,
@@ -458,8 +445,10 @@ static struct ib_srq *mthca_create_srq(struct ib_pd *pd,
 	if (pd->uobject) {
 		context = to_mucontext(pd->uobject->context);
 
-		if (ib_copy_from_udata(&ucmd, udata, sizeof ucmd))
-			return ERR_PTR(-EFAULT);
+		if (ib_copy_from_udata(&ucmd, udata, sizeof ucmd)) {
+			err = -EFAULT;
+			goto err_free;
+		}
 
 		err = mthca_map_user_db(to_mdev(pd->device), &context->uar,
 					context->db_tab, ucmd.db_index,
@@ -535,8 +524,10 @@ static struct ib_qp *mthca_create_qp(struct ib_pd *pd,
 		if (pd->uobject) {
 			context = to_mucontext(pd->uobject->context);
 
-			if (ib_copy_from_udata(&ucmd, udata, sizeof ucmd))
+			if (ib_copy_from_udata(&ucmd, udata, sizeof ucmd)) {
+				kfree(qp);
 				return ERR_PTR(-EFAULT);
+			}
 
 			err = mthca_map_user_db(to_mdev(pd->device), &context->uar,
 						context->db_tab,
@@ -783,24 +774,20 @@ static struct ib_mr *mthca_reg_phys_mr(struct ib_pd       *pd,
 	if ((*iova_start & ~PAGE_MASK) != (buffer_list[0].addr & ~PAGE_MASK))
 		return ERR_PTR(-EINVAL);
 
-	if (num_phys_buf > 1 &&
-	    ((buffer_list[0].addr + buffer_list[0].size) & ~PAGE_MASK))
-		return ERR_PTR(-EINVAL);
-
 	mask = 0;
 	total_size = 0;
 	for (i = 0; i < num_phys_buf; ++i) {
-		if (i != 0 && buffer_list[i].addr & ~PAGE_MASK)
-			return ERR_PTR(-EINVAL);
-		if (i != 0 && i != num_phys_buf - 1 &&
-		    (buffer_list[i].size & ~PAGE_MASK))
-			return ERR_PTR(-EINVAL);
+		if (i != 0)
+			mask |= buffer_list[i].addr;
+		if (i != num_phys_buf - 1)
+			mask |= buffer_list[i].addr + buffer_list[i].size;
 
 		total_size += buffer_list[i].size;
-		if (i > 0)
-			mask |= buffer_list[i].addr;
 	}
 
+	if (mask & ~PAGE_MASK)
+		return ERR_PTR(-EINVAL);
+
 	/* Find largest page shift we can use to cover buffers */
 	for (shift = PAGE_SHIFT; shift < 31; ++shift)
 		if (num_phys_buf > 1) {
@@ -1070,11 +1057,48 @@ static struct class_device_attribute *mthca_class_attributes[] = {
 	&class_device_attr_board_id
 };
 
+static int mthca_init_node_data(struct mthca_dev *dev)
+{
+	struct ib_smp *in_mad  = NULL;
+	struct ib_smp *out_mad = NULL;
+	int err = -ENOMEM;
+	u8 status;
+
+	in_mad  = kzalloc(sizeof *in_mad, GFP_KERNEL);
+	out_mad = kmalloc(sizeof *out_mad, GFP_KERNEL);
+	if (!in_mad || !out_mad)
+		goto out;
+
+	init_query_mad(in_mad);
+	in_mad->attr_id = IB_SMP_ATTR_NODE_INFO;
+
+	err = mthca_MAD_IFC(dev, 1, 1,
+			    1, NULL, NULL, in_mad, out_mad,
+			    &status);
+	if (err)
+		goto out;
+	if (status) {
+		err = -EINVAL;
+		goto out;
+	}
+
+	memcpy(&dev->ib_dev.node_guid, out_mad->data + 12, 8);
+
+out:
+	kfree(in_mad);
+	kfree(out_mad);
+	return err;
+}
+
 int mthca_register_device(struct mthca_dev *dev)
 {
 	int ret;
 	int i;
 
+	ret = mthca_init_node_data(dev);
+	if (ret)
+		return ret;
+
 	strlcpy(dev->ib_dev.name, "mthca%d", IB_DEVICE_NAME_MAX);
 	dev->ib_dev.owner                = THIS_MODULE;
 

drivers/infiniband/hw/mthca/mthca_qp.c

@@ -1434,7 +1434,7 @@ static int build_mlx_header(struct mthca_dev *dev, struct mthca_sqp *sqp,
 	u16 pkey;
 
 	ib_ud_header_init(256, /* assume a MAD */
-			  sqp->ud_header.grh_present,
+			  mthca_ah_grh_present(to_mah(wr->wr.ud.ah)),
 			  &sqp->ud_header);
 
 	err = mthca_read_ah(dev, to_mah(wr->wr.ud.ah), &sqp->ud_header);

drivers/infiniband/ulp/ipoib/ipoib.h

@@ -45,11 +45,11 @@
 #include <linux/config.h>
 #include <linux/kref.h>
 #include <linux/if_infiniband.h>
+#include <linux/mutex.h>
 
 #include <net/neighbour.h>
 
 #include <asm/atomic.h>
-#include <asm/semaphore.h>
 
 #include <rdma/ib_verbs.h>
 #include <rdma/ib_pack.h>
@@ -123,8 +123,8 @@ struct ipoib_dev_priv {
 
 	unsigned long flags;
 
-	struct semaphore mcast_mutex;
-	struct semaphore vlan_mutex;
+	struct mutex mcast_mutex;
+	struct mutex vlan_mutex;
 
 	struct rb_root  path_tree;
 	struct list_head path_list;

drivers/infiniband/ulp/ipoib/ipoib_ib.c

@@ -52,7 +52,7 @@ MODULE_PARM_DESC(data_debug_level,
 
 #define	IPOIB_OP_RECV	(1ul << 31)
 
-static DECLARE_MUTEX(pkey_sem);
+static DEFINE_MUTEX(pkey_mutex);
 
 struct ipoib_ah *ipoib_create_ah(struct net_device *dev,
 				 struct ib_pd *pd, struct ib_ah_attr *attr)
@@ -445,25 +445,16 @@ int ipoib_ib_dev_down(struct net_device *dev)
 
 	/* Shutdown the P_Key thread if still active */
 	if (!test_bit(IPOIB_PKEY_ASSIGNED, &priv->flags)) {
-		down(&pkey_sem);
+		mutex_lock(&pkey_mutex);
 		set_bit(IPOIB_PKEY_STOP, &priv->flags);
 		cancel_delayed_work(&priv->pkey_task);
-		up(&pkey_sem);
+		mutex_unlock(&pkey_mutex);
 		flush_workqueue(ipoib_workqueue);
 	}
 
 	ipoib_mcast_stop_thread(dev, 1);
-
-	/*
-	 * Flush the multicast groups first so we stop any multicast joins. The
-	 * completion thread may have already died and we may deadlock waiting
-	 * for the completion thread to finish some multicast joins.
-	 */
 	ipoib_mcast_dev_flush(dev);
 
-	/* Delete broadcast and local addresses since they will be recreated */
-	ipoib_mcast_dev_down(dev);
-
 	ipoib_flush_paths(dev);
 
 	return 0;
@@ -608,13 +599,13 @@ void ipoib_ib_dev_flush(void *_dev)
 	if (test_bit(IPOIB_FLAG_ADMIN_UP, &priv->flags))
 		ipoib_ib_dev_up(dev);
 
-	down(&priv->vlan_mutex);
+	mutex_lock(&priv->vlan_mutex);
 
 	/* Flush any child interfaces too */
 	list_for_each_entry(cpriv, &priv->child_intfs, list)
 		ipoib_ib_dev_flush(&cpriv->dev);
 
-	up(&priv->vlan_mutex);
+	mutex_unlock(&priv->vlan_mutex);
 }
 
 void ipoib_ib_dev_cleanup(struct net_device *dev)
@@ -624,9 +615,7 @@ void ipoib_ib_dev_cleanup(struct net_device *dev)
 	ipoib_dbg(priv, "cleaning up ib_dev\n");
 
 	ipoib_mcast_stop_thread(dev, 1);
-
-	/* Delete the broadcast address and the local address */
-	ipoib_mcast_dev_down(dev);
+	ipoib_mcast_dev_flush(dev);
 
 	ipoib_transport_dev_cleanup(dev);
 }
@@ -662,12 +651,12 @@ void ipoib_pkey_poll(void *dev_ptr)
 	if (test_bit(IPOIB_PKEY_ASSIGNED, &priv->flags))
 		ipoib_open(dev);
 	else {
-		down(&pkey_sem);
+		mutex_lock(&pkey_mutex);
 		if (!test_bit(IPOIB_PKEY_STOP, &priv->flags))
 			queue_delayed_work(ipoib_workqueue,
 					   &priv->pkey_task,
 					   HZ);
-		up(&pkey_sem);
+		mutex_unlock(&pkey_mutex);
 	}
 }
 
@@ -681,12 +670,12 @@ int ipoib_pkey_dev_delay_open(struct net_device *dev)
 
 	/* P_Key value not assigned yet - start polling */
 	if (!test_bit(IPOIB_PKEY_ASSIGNED, &priv->flags)) {
-		down(&pkey_sem);
+		mutex_lock(&pkey_mutex);
 		clear_bit(IPOIB_PKEY_STOP, &priv->flags);
 		queue_delayed_work(ipoib_workqueue,
 				   &priv->pkey_task,
 				   HZ);
-		up(&pkey_sem);
+		mutex_unlock(&pkey_mutex);
 		return 1;
 	}
 

drivers/infiniband/ulp/ipoib/ipoib_main.c

@@ -105,7 +105,7 @@ int ipoib_open(struct net_device *dev)
 		struct ipoib_dev_priv *cpriv;
 
 		/* Bring up any child interfaces too */
-		down(&priv->vlan_mutex);
+		mutex_lock(&priv->vlan_mutex);
 		list_for_each_entry(cpriv, &priv->child_intfs, list) {
 			int flags;
 
@@ -115,7 +115,7 @@ int ipoib_open(struct net_device *dev)
 
 			dev_change_flags(cpriv->dev, flags | IFF_UP);
 		}
-		up(&priv->vlan_mutex);
+		mutex_unlock(&priv->vlan_mutex);
 	}
 
 	netif_start_queue(dev);
@@ -140,7 +140,7 @@ static int ipoib_stop(struct net_device *dev)
 		struct ipoib_dev_priv *cpriv;
 
 		/* Bring down any child interfaces too */
-		down(&priv->vlan_mutex);
+		mutex_lock(&priv->vlan_mutex);
 		list_for_each_entry(cpriv, &priv->child_intfs, list) {
 			int flags;
 
@@ -150,7 +150,7 @@ static int ipoib_stop(struct net_device *dev)
 
 			dev_change_flags(cpriv->dev, flags & ~IFF_UP);
 		}
-		up(&priv->vlan_mutex);
+		mutex_unlock(&priv->vlan_mutex);
 	}
 
 	return 0;
@@ -892,8 +892,8 @@ static void ipoib_setup(struct net_device *dev)
 	spin_lock_init(&priv->lock);
 	spin_lock_init(&priv->tx_lock);
 
-	init_MUTEX(&priv->mcast_mutex);
-	init_MUTEX(&priv->vlan_mutex);
+	mutex_init(&priv->mcast_mutex);
+	mutex_init(&priv->vlan_mutex);
 
 	INIT_LIST_HEAD(&priv->path_list);
 	INIT_LIST_HEAD(&priv->child_intfs);

drivers/infiniband/ulp/ipoib/ipoib_multicast.c

@@ -55,7 +55,7 @@ MODULE_PARM_DESC(mcast_debug_level,
 		 "Enable multicast debug tracing if > 0");
 #endif
 
-static DECLARE_MUTEX(mcast_mutex);
+static DEFINE_MUTEX(mcast_mutex);
 
 /* Used for all multicast joins (broadcast, IPv4 mcast and IPv6 mcast) */
 struct ipoib_mcast {
@@ -97,8 +97,6 @@ static void ipoib_mcast_free(struct ipoib_mcast *mcast)
 	struct ipoib_dev_priv *priv = netdev_priv(dev);
 	struct ipoib_neigh *neigh, *tmp;
 	unsigned long flags;
-	LIST_HEAD(ah_list);
-	struct ipoib_ah *ah, *tah;
 
 	ipoib_dbg_mcast(netdev_priv(dev),
 			"deleting multicast group " IPOIB_GID_FMT "\n",
@@ -107,8 +105,14 @@ static void ipoib_mcast_free(struct ipoib_mcast *mcast)
 	spin_lock_irqsave(&priv->lock, flags);
 
 	list_for_each_entry_safe(neigh, tmp, &mcast->neigh_list, list) {
+		/*
+		 * It's safe to call ipoib_put_ah() inside priv->lock
+		 * here, because we know that mcast->ah will always
+		 * hold one more reference, so ipoib_put_ah() will
+		 * never do more than decrement the ref count.
+		 */
 		if (neigh->ah)
-			list_add_tail(&neigh->ah->list, &ah_list);
+			ipoib_put_ah(neigh->ah);
 		*to_ipoib_neigh(neigh->neighbour) = NULL;
 		neigh->neighbour->ops->destructor = NULL;
 		kfree(neigh);
@@ -116,9 +120,6 @@ static void ipoib_mcast_free(struct ipoib_mcast *mcast)
 
 	spin_unlock_irqrestore(&priv->lock, flags);
 
-	list_for_each_entry_safe(ah, tah, &ah_list, list)
-		ipoib_put_ah(ah);
-
 	if (mcast->ah)
 		ipoib_put_ah(mcast->ah);
 
@@ -384,10 +385,10 @@ static void ipoib_mcast_join_complete(int status,
 
 	if (!status && !ipoib_mcast_join_finish(mcast, mcmember)) {
 		mcast->backoff = 1;
-		down(&mcast_mutex);
+		mutex_lock(&mcast_mutex);
 		if (test_bit(IPOIB_MCAST_RUN, &priv->flags))
 			queue_work(ipoib_workqueue, &priv->mcast_task);
-		up(&mcast_mutex);
+		mutex_unlock(&mcast_mutex);
 		complete(&mcast->done);
 		return;
 	}
@@ -417,7 +418,7 @@ static void ipoib_mcast_join_complete(int status,
 
 	mcast->query = NULL;
 
-	down(&mcast_mutex);
+	mutex_lock(&mcast_mutex);
 	if (test_bit(IPOIB_MCAST_RUN, &priv->flags)) {
 		if (status == -ETIMEDOUT)
 			queue_work(ipoib_workqueue, &priv->mcast_task);
@@ -426,7 +427,7 @@ static void ipoib_mcast_join_complete(int status,
 					   mcast->backoff * HZ);
 	} else
 		complete(&mcast->done);
-	up(&mcast_mutex);
+	mutex_unlock(&mcast_mutex);
 
 	return;
 }
@@ -481,12 +482,12 @@ static void ipoib_mcast_join(struct net_device *dev, struct ipoib_mcast *mcast,
 		if (mcast->backoff > IPOIB_MAX_BACKOFF_SECONDS)
 			mcast->backoff = IPOIB_MAX_BACKOFF_SECONDS;
 
-		down(&mcast_mutex);
+		mutex_lock(&mcast_mutex);
 		if (test_bit(IPOIB_MCAST_RUN, &priv->flags))
 			queue_delayed_work(ipoib_workqueue,
 					   &priv->mcast_task,
 					   mcast->backoff * HZ);
-		up(&mcast_mutex);
+		mutex_unlock(&mcast_mutex);
 	} else
 		mcast->query_id = ret;
 }
@@ -519,11 +520,11 @@ void ipoib_mcast_join_task(void *dev_ptr)
 		priv->broadcast = ipoib_mcast_alloc(dev, 1);
 		if (!priv->broadcast) {
 			ipoib_warn(priv, "failed to allocate broadcast group\n");
-			down(&mcast_mutex);
+			mutex_lock(&mcast_mutex);
 			if (test_bit(IPOIB_MCAST_RUN, &priv->flags))
 				queue_delayed_work(ipoib_workqueue,
 						   &priv->mcast_task, HZ);
-			up(&mcast_mutex);
+			mutex_unlock(&mcast_mutex);
 			return;
 		}
 
@@ -579,10 +580,10 @@ int ipoib_mcast_start_thread(struct net_device *dev)
 
 	ipoib_dbg_mcast(priv, "starting multicast thread\n");
 
-	down(&mcast_mutex);
+	mutex_lock(&mcast_mutex);
 	if (!test_and_set_bit(IPOIB_MCAST_RUN, &priv->flags))
 		queue_work(ipoib_workqueue, &priv->mcast_task);
-	up(&mcast_mutex);
+	mutex_unlock(&mcast_mutex);
 
 	return 0;
 }
@@ -594,10 +595,10 @@ int ipoib_mcast_stop_thread(struct net_device *dev, int flush)
 
 	ipoib_dbg_mcast(priv, "stopping multicast thread\n");
 
-	down(&mcast_mutex);
+	mutex_lock(&mcast_mutex);
 	clear_bit(IPOIB_MCAST_RUN, &priv->flags);
 	cancel_delayed_work(&priv->mcast_task);
-	up(&mcast_mutex);
+	mutex_unlock(&mcast_mutex);
 
 	if (flush)
 		flush_workqueue(ipoib_workqueue);
@@ -741,48 +742,23 @@ void ipoib_mcast_dev_flush(struct net_device *dev)
 {
 	struct ipoib_dev_priv *priv = netdev_priv(dev);
 	LIST_HEAD(remove_list);
-	struct ipoib_mcast *mcast, *tmcast, *nmcast;
+	struct ipoib_mcast *mcast, *tmcast;
 	unsigned long flags;
 
 	ipoib_dbg_mcast(priv, "flushing multicast list\n");
 
 	spin_lock_irqsave(&priv->lock, flags);
-	list_for_each_entry_safe(mcast, tmcast, &priv->multicast_list, list) {
-		nmcast = ipoib_mcast_alloc(dev, 0);
-		if (nmcast) {
-			nmcast->flags =
-				mcast->flags & (1 << IPOIB_MCAST_FLAG_SENDONLY);
 
-			nmcast->mcmember.mgid = mcast->mcmember.mgid;
-
-			/* Add the new group in before the to-be-destroyed group */
-			list_add_tail(&nmcast->list, &mcast->list);
-			list_del_init(&mcast->list);
-
-			rb_replace_node(&mcast->rb_node, &nmcast->rb_node,
-					&priv->multicast_tree);
-
-			list_add_tail(&mcast->list, &remove_list);
-		} else {
-			ipoib_warn(priv, "could not reallocate multicast group "
-				   IPOIB_GID_FMT "\n",
-				   IPOIB_GID_ARG(mcast->mcmember.mgid));
-		}
+	list_for_each_entry_safe(mcast, tmcast, &priv->multicast_list, list) {
+		list_del(&mcast->list);
+		rb_erase(&mcast->rb_node, &priv->multicast_tree);
+		list_add_tail(&mcast->list, &remove_list);
 	}
 
 	if (priv->broadcast) {
-		nmcast = ipoib_mcast_alloc(dev, 0);
-		if (nmcast) {
-			nmcast->mcmember.mgid = priv->broadcast->mcmember.mgid;
-
-			rb_replace_node(&priv->broadcast->rb_node,
-					&nmcast->rb_node,
-					&priv->multicast_tree);
-
-			list_add_tail(&priv->broadcast->list, &remove_list);
-		}
-
-		priv->broadcast = nmcast;
+ 		rb_erase(&priv->broadcast->rb_node, &priv->multicast_tree);
+		list_add_tail(&priv->broadcast->list, &remove_list);
+		priv->broadcast = NULL;
 	}
 
 	spin_unlock_irqrestore(&priv->lock, flags);
@@ -793,24 +769,6 @@ void ipoib_mcast_dev_flush(struct net_device *dev)
 	}
 }
 
-void ipoib_mcast_dev_down(struct net_device *dev)
-{
-	struct ipoib_dev_priv *priv = netdev_priv(dev);
-	unsigned long flags;
-
-	/* Delete broadcast since it will be recreated */
-	if (priv->broadcast) {
-		ipoib_dbg_mcast(priv, "deleting broadcast group\n");
-
-		spin_lock_irqsave(&priv->lock, flags);
-		rb_erase(&priv->broadcast->rb_node, &priv->multicast_tree);
-		spin_unlock_irqrestore(&priv->lock, flags);
-		ipoib_mcast_leave(dev, priv->broadcast);
-		ipoib_mcast_free(priv->broadcast);
-		priv->broadcast = NULL;
-	}
-}
-
 void ipoib_mcast_restart_task(void *dev_ptr)
 {
 	struct net_device *dev = dev_ptr;
@@ -824,7 +782,8 @@ void ipoib_mcast_restart_task(void *dev_ptr)
 
 	ipoib_mcast_stop_thread(dev, 0);
 
-	spin_lock_irqsave(&priv->lock, flags);
+	spin_lock_irqsave(&dev->xmit_lock, flags);
+	spin_lock(&priv->lock);
 
 	/*
 	 * Unfortunately, the networking core only gives us a list of all of
@@ -896,7 +855,9 @@ void ipoib_mcast_restart_task(void *dev_ptr)
 			list_add_tail(&mcast->list, &remove_list);
 		}
 	}
-	spin_unlock_irqrestore(&priv->lock, flags);
+
+	spin_unlock(&priv->lock);
+	spin_unlock_irqrestore(&dev->xmit_lock, flags);
 
 	/* We have to cancel outside of the spinlock */
 	list_for_each_entry_safe(mcast, tmcast, &remove_list, list) {

drivers/infiniband/ulp/ipoib/ipoib_verbs.c

@@ -65,9 +65,9 @@ int ipoib_mcast_attach(struct net_device *dev, u16 mlid, union ib_gid *mgid)
 	}
 
 	/* attach QP to multicast group */
-	down(&priv->mcast_mutex);
+	mutex_lock(&priv->mcast_mutex);
 	ret = ib_attach_mcast(priv->qp, mgid, mlid);
-	up(&priv->mcast_mutex);
+	mutex_unlock(&priv->mcast_mutex);
 	if (ret)
 		ipoib_warn(priv, "failed to attach to multicast group, ret = %d\n", ret);
 
@@ -81,9 +81,9 @@ int ipoib_mcast_detach(struct net_device *dev, u16 mlid, union ib_gid *mgid)
 	struct ipoib_dev_priv *priv = netdev_priv(dev);
 	int ret;
 
-	down(&priv->mcast_mutex);
+	mutex_lock(&priv->mcast_mutex);
 	ret = ib_detach_mcast(priv->qp, mgid, mlid);
-	up(&priv->mcast_mutex);
+	mutex_unlock(&priv->mcast_mutex);
 	if (ret)
 		ipoib_warn(priv, "ib_detach_mcast failed (result = %d)\n", ret);
 

drivers/infiniband/ulp/ipoib/ipoib_vlan.c

@@ -63,7 +63,7 @@ int ipoib_vlan_add(struct net_device *pdev, unsigned short pkey)
 
 	ppriv = netdev_priv(pdev);
 
-	down(&ppriv->vlan_mutex);
+	mutex_lock(&ppriv->vlan_mutex);
 
 	/*
 	 * First ensure this isn't a duplicate. We check the parent device and
@@ -124,7 +124,7 @@ int ipoib_vlan_add(struct net_device *pdev, unsigned short pkey)
 
 	list_add_tail(&priv->list, &ppriv->child_intfs);
 
-	up(&ppriv->vlan_mutex);
+	mutex_unlock(&ppriv->vlan_mutex);
 
 	return 0;
 
@@ -139,7 +139,7 @@ int ipoib_vlan_add(struct net_device *pdev, unsigned short pkey)
 	free_netdev(priv->dev);
 
 err:
-	up(&ppriv->vlan_mutex);
+	mutex_unlock(&ppriv->vlan_mutex);
 	return result;
 }
 
@@ -153,7 +153,7 @@ int ipoib_vlan_delete(struct net_device *pdev, unsigned short pkey)
 
 	ppriv = netdev_priv(pdev);
 
-	down(&ppriv->vlan_mutex);
+	mutex_lock(&ppriv->vlan_mutex);
 	list_for_each_entry_safe(priv, tpriv, &ppriv->child_intfs, list) {
 		if (priv->pkey == pkey) {
 			unregister_netdev(priv->dev);
@@ -167,7 +167,7 @@ int ipoib_vlan_delete(struct net_device *pdev, unsigned short pkey)
 			break;
 		}
 	}
-	up(&ppriv->vlan_mutex);
+	mutex_unlock(&ppriv->vlan_mutex);
 
 	return ret;
 }

drivers/infiniband/ulp/srp/ib_srp.c

@@ -1516,8 +1516,7 @@ static ssize_t show_port(struct class_device *class_dev, char *buf)
 
 static CLASS_DEVICE_ATTR(port, S_IRUGO, show_port, NULL);
 
-static struct srp_host *srp_add_port(struct ib_device *device,
-				     __be64 node_guid, u8 port)
+static struct srp_host *srp_add_port(struct ib_device *device, u8 port)
 {
 	struct srp_host *host;
 
@@ -1532,7 +1531,7 @@ static struct srp_host *srp_add_port(struct ib_device *device,
 	host->port = port;
 
 	host->initiator_port_id[7] = port;
-	memcpy(host->initiator_port_id + 8, &node_guid, 8);
+	memcpy(host->initiator_port_id + 8, &device->node_guid, 8);
 
 	host->pd   = ib_alloc_pd(device);
 	if (IS_ERR(host->pd))
@@ -1580,22 +1579,11 @@ static void srp_add_one(struct ib_device *device)
 {
 	struct list_head *dev_list;
 	struct srp_host *host;
-	struct ib_device_attr *dev_attr;
 	int s, e, p;
 
-	dev_attr = kmalloc(sizeof *dev_attr, GFP_KERNEL);
-	if (!dev_attr)
-		return;
-
-	if (ib_query_device(device, dev_attr)) {
-		printk(KERN_WARNING PFX "Couldn't query node GUID for %s.\n",
-		       device->name);
-		goto out;
-	}
-
 	dev_list = kmalloc(sizeof *dev_list, GFP_KERNEL);
 	if (!dev_list)
-		goto out;
+		return;
 
 	INIT_LIST_HEAD(dev_list);
 
@@ -1608,15 +1596,12 @@ static void srp_add_one(struct ib_device *device)
 	}
 
 	for (p = s; p <= e; ++p) {
-		host = srp_add_port(device, dev_attr->node_guid, p);
+		host = srp_add_port(device, p);
 		if (host)
 			list_add_tail(&host->list, dev_list);
 	}
 
 	ib_set_client_data(device, &srp_client, dev_list);
-
-out:
-	kfree(dev_attr);
 }
 
 static void srp_remove_one(struct ib_device *device)

include/rdma/ib_verbs.h

@@ -88,7 +88,6 @@ enum ib_atomic_cap {
 
 struct ib_device_attr {
 	u64			fw_ver;
-	__be64			node_guid;
 	__be64			sys_image_guid;
 	u64			max_mr_size;
 	u64			page_size_cap;
@@ -951,6 +950,7 @@ struct ib_device {
 	u64			     uverbs_cmd_mask;
 	int			     uverbs_abi_ver;
 
+	__be64			     node_guid;
 	u8                           node_type;
 	u8                           phys_port_cnt;
 };