Merge tag 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/rdma/rdma

Pull rdma fixes from Jason Gunthorpe:
 "The usual collection of driver bug fixes, and a few regressions from
  the merge window. Nothing particularly worrisome.

   - Various missed memory frees and error unwind bugs

   - Fix regressions in a few iwarp drivers from 5.4 patches

   - A few regressions added in past kernels

   - Squash a number of races in mlx5 ODP code"

* tag 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/rdma/rdma:
  RDMA/mlx5: Add missing synchronize_srcu() for MW cases
  RDMA/mlx5: Put live in the correct place for ODP MRs
  RDMA/mlx5: Order num_pending_prefetch properly with synchronize_srcu
  RDMA/odp: Lift umem_mutex out of ib_umem_odp_unmap_dma_pages()
  RDMA/mlx5: Fix a race with mlx5_ib_update_xlt on an implicit MR
  RDMA/mlx5: Do not allow rereg of a ODP MR
  IB/core: Fix wrong iterating on ports
  RDMA/nldev: Reshuffle the code to avoid need to rebind QP in error path
  RDMA/cxgb4: Do not dma memory off of the stack
  RDMA/cm: Fix memory leak in cm_add/remove_one
  RDMA/core: Fix an error handling path in 'res_get_common_doit()'
  RDMA/i40iw: Associate ibdev to netdev before IB device registration
  RDMA/iwcm: Fix a lock inversion issue
  RDMA/iw_cxgb4: fix SRQ access from dump_qp()
  RDMA/hfi1: Prevent memory leak in sdma_init
  RDMA/core: Fix use after free and refcnt leak on ndev in_device in iwarp_query_port
  RDMA/siw: Fix serialization issue in write_space()
  RDMA/vmw_pvrdma: Free SRQ only once

drivers/infiniband/core/cm.c

@@ -4399,6 +4399,7 @@ static void cm_add_one(struct ib_device *ib_device)
 error1:
 	port_modify.set_port_cap_mask = 0;
 	port_modify.clr_port_cap_mask = IB_PORT_CM_SUP;
+	kfree(port);
 	while (--i) {
 		if (!rdma_cap_ib_cm(ib_device, i))
 			continue;
@@ -4407,6 +4408,7 @@ static void cm_add_one(struct ib_device *ib_device)
 		ib_modify_port(ib_device, port->port_num, 0, &port_modify);
 		ib_unregister_mad_agent(port->mad_agent);
 		cm_remove_port_fs(port);
+		kfree(port);
 	}
 free:
 	kfree(cm_dev);
@@ -4460,6 +4462,7 @@ static void cm_remove_one(struct ib_device *ib_device, void *client_data)
 		spin_unlock_irq(&cm.state_lock);
 		ib_unregister_mad_agent(cur_mad_agent);
 		cm_remove_port_fs(port);
+		kfree(port);
 	}
 
 	kfree(cm_dev);

drivers/infiniband/core/cma.c

@@ -2396,9 +2396,10 @@ static int iw_conn_req_handler(struct iw_cm_id *cm_id,
 		conn_id->cm_id.iw = NULL;
 		cma_exch(conn_id, RDMA_CM_DESTROYING);
 		mutex_unlock(&conn_id->handler_mutex);
+		mutex_unlock(&listen_id->handler_mutex);
 		cma_deref_id(conn_id);
 		rdma_destroy_id(&conn_id->id);
-		goto out;
+		return ret;
 	}
 
 	mutex_unlock(&conn_id->handler_mutex);

drivers/infiniband/core/device.c

@@ -1987,8 +1987,6 @@ static int iw_query_port(struct ib_device *device,
 	if (!netdev)
 		return -ENODEV;
 
-	dev_put(netdev);
-
 	port_attr->max_mtu = IB_MTU_4096;
 	port_attr->active_mtu = ib_mtu_int_to_enum(netdev->mtu);
 
@@ -1996,19 +1994,22 @@ static int iw_query_port(struct ib_device *device,
 		port_attr->state = IB_PORT_DOWN;
 		port_attr->phys_state = IB_PORT_PHYS_STATE_DISABLED;
 	} else {
-		inetdev = in_dev_get(netdev);
+		rcu_read_lock();
+		inetdev = __in_dev_get_rcu(netdev);
 
 		if (inetdev && inetdev->ifa_list) {
 			port_attr->state = IB_PORT_ACTIVE;
 			port_attr->phys_state = IB_PORT_PHYS_STATE_LINK_UP;
-			in_dev_put(inetdev);
 		} else {
 			port_attr->state = IB_PORT_INIT;
 			port_attr->phys_state =
 				IB_PORT_PHYS_STATE_PORT_CONFIGURATION_TRAINING;
 		}
+
+		rcu_read_unlock();
 	}
 
+	dev_put(netdev);
 	err = device->ops.query_port(device, port_num, port_attr);
 	if (err)
 		return err;

drivers/infiniband/core/nldev.c

@@ -1230,7 +1230,7 @@ static int res_get_common_doit(struct sk_buff *skb, struct nlmsghdr *nlh,
 	msg = nlmsg_new(NLMSG_DEFAULT_SIZE, GFP_KERNEL);
 	if (!msg) {
 		ret = -ENOMEM;
-		goto err;
+		goto err_get;
 	}
 
 	nlh = nlmsg_put(msg, NETLINK_CB(skb).portid, nlh->nlmsg_seq,
@@ -1787,10 +1787,6 @@ static int nldev_stat_del_doit(struct sk_buff *skb, struct nlmsghdr *nlh,
 
 	cntn = nla_get_u32(tb[RDMA_NLDEV_ATTR_STAT_COUNTER_ID]);
 	qpn = nla_get_u32(tb[RDMA_NLDEV_ATTR_RES_LQPN]);
-	ret = rdma_counter_unbind_qpn(device, port, qpn, cntn);
-	if (ret)
-		goto err_unbind;
-
 	if (fill_nldev_handle(msg, device) ||
 	    nla_put_u32(msg, RDMA_NLDEV_ATTR_PORT_INDEX, port) ||
 	    nla_put_u32(msg, RDMA_NLDEV_ATTR_STAT_COUNTER_ID, cntn) ||
@@ -1799,13 +1795,15 @@ static int nldev_stat_del_doit(struct sk_buff *skb, struct nlmsghdr *nlh,
 		goto err_fill;
 	}
 
+	ret = rdma_counter_unbind_qpn(device, port, qpn, cntn);
+	if (ret)
+		goto err_fill;
+
 	nlmsg_end(msg, nlh);
 	ib_device_put(device);
 	return rdma_nl_unicast(sock_net(skb->sk), msg, NETLINK_CB(skb).portid);
 
 err_fill:
-	rdma_counter_bind_qpn(device, port, qpn, cntn);
-err_unbind:
 	nlmsg_free(msg);
 err:
 	ib_device_put(device);

drivers/infiniband/core/security.c

@@ -426,7 +426,7 @@ int ib_create_qp_security(struct ib_qp *qp, struct ib_device *dev)
 	int ret;
 
 	rdma_for_each_port (dev, i) {
-		is_ib = rdma_protocol_ib(dev, i++);
+		is_ib = rdma_protocol_ib(dev, i);
 		if (is_ib)
 			break;
 	}

drivers/infiniband/core/umem_odp.c

@@ -451,8 +451,10 @@ void ib_umem_odp_release(struct ib_umem_odp *umem_odp)
 	 * that the hardware will not attempt to access the MR any more.
 	 */
 	if (!umem_odp->is_implicit_odp) {
+		mutex_lock(&umem_odp->umem_mutex);
 		ib_umem_odp_unmap_dma_pages(umem_odp, ib_umem_start(umem_odp),
 					    ib_umem_end(umem_odp));
+		mutex_unlock(&umem_odp->umem_mutex);
 		kvfree(umem_odp->dma_list);
 		kvfree(umem_odp->page_list);
 	}
@@ -719,6 +721,8 @@ void ib_umem_odp_unmap_dma_pages(struct ib_umem_odp *umem_odp, u64 virt,
 	u64 addr;
 	struct ib_device *dev = umem_odp->umem.ibdev;
 
+	lockdep_assert_held(&umem_odp->umem_mutex);
+
 	virt = max_t(u64, virt, ib_umem_start(umem_odp));
 	bound = min_t(u64, bound, ib_umem_end(umem_odp));
 	/* Note that during the run of this function, the
@@ -726,7 +730,6 @@ void ib_umem_odp_unmap_dma_pages(struct ib_umem_odp *umem_odp, u64 virt,
 	 * faults from completion. We might be racing with other
 	 * invalidations, so we must make sure we free each page only
 	 * once. */
-	mutex_lock(&umem_odp->umem_mutex);
 	for (addr = virt; addr < bound; addr += BIT(umem_odp->page_shift)) {
 		idx = (addr - ib_umem_start(umem_odp)) >> umem_odp->page_shift;
 		if (umem_odp->page_list[idx]) {
@@ -757,7 +760,6 @@ void ib_umem_odp_unmap_dma_pages(struct ib_umem_odp *umem_odp, u64 virt,
 			umem_odp->npages--;
 		}
 	}
-	mutex_unlock(&umem_odp->umem_mutex);
 }
 EXPORT_SYMBOL(ib_umem_odp_unmap_dma_pages);
 

drivers/infiniband/hw/cxgb4/device.c

@@ -242,10 +242,13 @@ static void set_ep_sin6_addrs(struct c4iw_ep *ep,
 	}
 }
 
-static int dump_qp(struct c4iw_qp *qp, struct c4iw_debugfs_data *qpd)
+static int dump_qp(unsigned long id, struct c4iw_qp *qp,
+		   struct c4iw_debugfs_data *qpd)
 {
 	int space;
 	int cc;
+	if (id != qp->wq.sq.qid)
+		return 0;
 
 	space = qpd->bufsize - qpd->pos - 1;
 	if (space == 0)
@@ -350,7 +353,7 @@ static int qp_open(struct inode *inode, struct file *file)
 
 	xa_lock_irq(&qpd->devp->qps);
 	xa_for_each(&qpd->devp->qps, index, qp)
-		dump_qp(qp, qpd);
+		dump_qp(index, qp, qpd);
 	xa_unlock_irq(&qpd->devp->qps);
 
 	qpd->buf[qpd->pos++] = 0;

drivers/infiniband/hw/cxgb4/mem.c

@@ -275,13 +275,17 @@ static int write_tpt_entry(struct c4iw_rdev *rdev, u32 reset_tpt_entry,
 			   struct sk_buff *skb, struct c4iw_wr_wait *wr_waitp)
 {
 	int err;
-	struct fw_ri_tpte tpt;
+	struct fw_ri_tpte *tpt;
 	u32 stag_idx;
 	static atomic_t key;
 
 	if (c4iw_fatal_error(rdev))
 		return -EIO;
 
+	tpt = kmalloc(sizeof(*tpt), GFP_KERNEL);
+	if (!tpt)
+		return -ENOMEM;
+
 	stag_state = stag_state > 0;
 	stag_idx = (*stag) >> 8;
 
@@ -291,6 +295,7 @@ static int write_tpt_entry(struct c4iw_rdev *rdev, u32 reset_tpt_entry,
 			mutex_lock(&rdev->stats.lock);
 			rdev->stats.stag.fail++;
 			mutex_unlock(&rdev->stats.lock);
+			kfree(tpt);
 			return -ENOMEM;
 		}
 		mutex_lock(&rdev->stats.lock);
@@ -305,28 +310,28 @@ static int write_tpt_entry(struct c4iw_rdev *rdev, u32 reset_tpt_entry,
 
 	/* write TPT entry */
 	if (reset_tpt_entry)
-		memset(&tpt, 0, sizeof(tpt));
+		memset(tpt, 0, sizeof(*tpt));
 	else {
-		tpt.valid_to_pdid = cpu_to_be32(FW_RI_TPTE_VALID_F |
+		tpt->valid_to_pdid = cpu_to_be32(FW_RI_TPTE_VALID_F |
 			FW_RI_TPTE_STAGKEY_V((*stag & FW_RI_TPTE_STAGKEY_M)) |
 			FW_RI_TPTE_STAGSTATE_V(stag_state) |
 			FW_RI_TPTE_STAGTYPE_V(type) | FW_RI_TPTE_PDID_V(pdid));
-		tpt.locread_to_qpid = cpu_to_be32(FW_RI_TPTE_PERM_V(perm) |
+		tpt->locread_to_qpid = cpu_to_be32(FW_RI_TPTE_PERM_V(perm) |
 			(bind_enabled ? FW_RI_TPTE_MWBINDEN_F : 0) |
 			FW_RI_TPTE_ADDRTYPE_V((zbva ? FW_RI_ZERO_BASED_TO :
 						      FW_RI_VA_BASED_TO))|
 			FW_RI_TPTE_PS_V(page_size));
-		tpt.nosnoop_pbladdr = !pbl_size ? 0 : cpu_to_be32(
+		tpt->nosnoop_pbladdr = !pbl_size ? 0 : cpu_to_be32(
 			FW_RI_TPTE_PBLADDR_V(PBL_OFF(rdev, pbl_addr)>>3));
-		tpt.len_lo = cpu_to_be32((u32)(len & 0xffffffffUL));
-		tpt.va_hi = cpu_to_be32((u32)(to >> 32));
-		tpt.va_lo_fbo = cpu_to_be32((u32)(to & 0xffffffffUL));
-		tpt.dca_mwbcnt_pstag = cpu_to_be32(0);
-		tpt.len_hi = cpu_to_be32((u32)(len >> 32));
+		tpt->len_lo = cpu_to_be32((u32)(len & 0xffffffffUL));
+		tpt->va_hi = cpu_to_be32((u32)(to >> 32));
+		tpt->va_lo_fbo = cpu_to_be32((u32)(to & 0xffffffffUL));
+		tpt->dca_mwbcnt_pstag = cpu_to_be32(0);
+		tpt->len_hi = cpu_to_be32((u32)(len >> 32));
 	}
 	err = write_adapter_mem(rdev, stag_idx +
 				(rdev->lldi.vr->stag.start >> 5),
-				sizeof(tpt), &tpt, skb, wr_waitp);
+				sizeof(*tpt), tpt, skb, wr_waitp);
 
 	if (reset_tpt_entry) {
 		c4iw_put_resource(&rdev->resource.tpt_table, stag_idx);
@@ -334,6 +339,7 @@ static int write_tpt_entry(struct c4iw_rdev *rdev, u32 reset_tpt_entry,
 		rdev->stats.stag.cur -= 32;
 		mutex_unlock(&rdev->stats.lock);
 	}
+	kfree(tpt);
 	return err;
 }
 

drivers/infiniband/hw/cxgb4/qp.c

@@ -2737,15 +2737,11 @@ int c4iw_create_srq(struct ib_srq *ib_srq, struct ib_srq_init_attr *attrs,
 	if (CHELSIO_CHIP_VERSION(rhp->rdev.lldi.adapter_type) > CHELSIO_T6)
 		srq->flags = T4_SRQ_LIMIT_SUPPORT;
 
-	ret = xa_insert_irq(&rhp->qps, srq->wq.qid, srq, GFP_KERNEL);
-	if (ret)
-		goto err_free_queue;
-
 	if (udata) {
 		srq_key_mm = kmalloc(sizeof(*srq_key_mm), GFP_KERNEL);
 		if (!srq_key_mm) {
 			ret = -ENOMEM;
-			goto err_remove_handle;
+			goto err_free_queue;
 		}
 		srq_db_key_mm = kmalloc(sizeof(*srq_db_key_mm), GFP_KERNEL);
 		if (!srq_db_key_mm) {
@@ -2789,8 +2785,6 @@ int c4iw_create_srq(struct ib_srq *ib_srq, struct ib_srq_init_attr *attrs,
 	kfree(srq_db_key_mm);
 err_free_srq_key_mm:
 	kfree(srq_key_mm);
-err_remove_handle:
-	xa_erase_irq(&rhp->qps, srq->wq.qid);
 err_free_queue:
 	free_srq_queue(srq, ucontext ? &ucontext->uctx : &rhp->rdev.uctx,
 		       srq->wr_waitp);
@@ -2813,8 +2807,6 @@ void c4iw_destroy_srq(struct ib_srq *ibsrq, struct ib_udata *udata)
 	rhp = srq->rhp;
 
 	pr_debug("%s id %d\n", __func__, srq->wq.qid);
-
-	xa_erase_irq(&rhp->qps, srq->wq.qid);
 	ucontext = rdma_udata_to_drv_context(udata, struct c4iw_ucontext,
 					     ibucontext);
 	free_srq_queue(srq, ucontext ? &ucontext->uctx : &rhp->rdev.uctx,

drivers/infiniband/hw/hfi1/sdma.c

@@ -1526,8 +1526,11 @@ int sdma_init(struct hfi1_devdata *dd, u8 port)
 	}
 
 	ret = rhashtable_init(tmp_sdma_rht, &sdma_rht_params);
-	if (ret < 0)
+	if (ret < 0) {
+		kfree(tmp_sdma_rht);
 		goto bail;
+	}
+
 	dd->sdma_rht = tmp_sdma_rht;
 
 	dd_dev_info(dd, "SDMA num_sdma: %u\n", dd->num_sdma);

drivers/infiniband/hw/i40iw/i40iw_verbs.c

@@ -2773,6 +2773,10 @@ int i40iw_register_rdma_device(struct i40iw_device *iwdev)
 		return -ENOMEM;
 	iwibdev = iwdev->iwibdev;
 	rdma_set_device_sysfs_group(&iwibdev->ibdev, &i40iw_attr_group);
+	ret = ib_device_set_netdev(&iwibdev->ibdev, iwdev->netdev, 1);
+	if (ret)
+		goto error;
+
 	ret = ib_register_device(&iwibdev->ibdev, "i40iw%d");
 	if (ret)
 		goto error;

drivers/infiniband/hw/mlx5/devx.c

@@ -1298,29 +1298,6 @@ static int devx_handle_mkey_create(struct mlx5_ib_dev *dev,
 	return 0;
 }
 
-static void devx_free_indirect_mkey(struct rcu_head *rcu)
-{
-	kfree(container_of(rcu, struct devx_obj, devx_mr.rcu));
-}
-
-/* This function to delete from the radix tree needs to be called before
- * destroying the underlying mkey. Otherwise a race might occur in case that
- * other thread will get the same mkey before this one will be deleted,
- * in that case it will fail via inserting to the tree its own data.
- *
- * Note:
- * An error in the destroy is not expected unless there is some other indirect
- * mkey which points to this one. In a kernel cleanup flow it will be just
- * destroyed in the iterative destruction call. In a user flow, in case
- * the application didn't close in the expected order it's its own problem,
- * the mkey won't be part of the tree, in both cases the kernel is safe.
- */
-static void devx_cleanup_mkey(struct devx_obj *obj)
-{
-	xa_erase(&obj->ib_dev->mdev->priv.mkey_table,
-		 mlx5_base_mkey(obj->devx_mr.mmkey.key));
-}
-
 static void devx_cleanup_subscription(struct mlx5_ib_dev *dev,
 				      struct devx_event_subscription *sub)
 {
@@ -1362,8 +1339,16 @@ static int devx_obj_cleanup(struct ib_uobject *uobject,
 	int ret;
 
 	dev = mlx5_udata_to_mdev(&attrs->driver_udata);
-	if (obj->flags & DEVX_OBJ_FLAGS_INDIRECT_MKEY)
-		devx_cleanup_mkey(obj);
+	if (obj->flags & DEVX_OBJ_FLAGS_INDIRECT_MKEY) {
+		/*
+		 * The pagefault_single_data_segment() does commands against
+		 * the mmkey, we must wait for that to stop before freeing the
+		 * mkey, as another allocation could get the same mkey #.
+		 */
+		xa_erase(&obj->ib_dev->mdev->priv.mkey_table,
+			 mlx5_base_mkey(obj->devx_mr.mmkey.key));
+		synchronize_srcu(&dev->mr_srcu);
+	}
 
 	if (obj->flags & DEVX_OBJ_FLAGS_DCT)
 		ret = mlx5_core_destroy_dct(obj->ib_dev->mdev, &obj->core_dct);
@@ -1382,12 +1367,6 @@ static int devx_obj_cleanup(struct ib_uobject *uobject,
 		devx_cleanup_subscription(dev, sub_entry);
 	mutex_unlock(&devx_event_table->event_xa_lock);
 
-	if (obj->flags & DEVX_OBJ_FLAGS_INDIRECT_MKEY) {
-		call_srcu(&dev->mr_srcu, &obj->devx_mr.rcu,
-			  devx_free_indirect_mkey);
-		return ret;
-	}
-
 	kfree(obj);
 	return ret;
 }
@@ -1491,26 +1470,21 @@ static int UVERBS_HANDLER(MLX5_IB_METHOD_DEVX_OBJ_CREATE)(
 				   &obj_id);
 	WARN_ON(obj->dinlen > MLX5_MAX_DESTROY_INBOX_SIZE_DW * sizeof(u32));
 
-	if (obj->flags & DEVX_OBJ_FLAGS_INDIRECT_MKEY) {
-		err = devx_handle_mkey_indirect(obj, dev, cmd_in, cmd_out);
-		if (err)
-			goto obj_destroy;
-	}
-
 	err = uverbs_copy_to(attrs, MLX5_IB_ATTR_DEVX_OBJ_CREATE_CMD_OUT, cmd_out, cmd_out_len);
 	if (err)
-		goto err_copy;
+		goto obj_destroy;
 
 	if (opcode == MLX5_CMD_OP_CREATE_GENERAL_OBJECT)
 		obj_type = MLX5_GET(general_obj_in_cmd_hdr, cmd_in, obj_type);
-
 	obj->obj_id = get_enc_obj_id(opcode | obj_type << 16, obj_id);
 
+	if (obj->flags & DEVX_OBJ_FLAGS_INDIRECT_MKEY) {
+		err = devx_handle_mkey_indirect(obj, dev, cmd_in, cmd_out);
+		if (err)
+			goto obj_destroy;
+	}
 	return 0;
 
-err_copy:
-	if (obj->flags & DEVX_OBJ_FLAGS_INDIRECT_MKEY)
-		devx_cleanup_mkey(obj);
 obj_destroy:
 	if (obj->flags & DEVX_OBJ_FLAGS_DCT)
 		mlx5_core_destroy_dct(obj->ib_dev->mdev, &obj->core_dct);

drivers/infiniband/hw/mlx5/mlx5_ib.h

@@ -606,7 +606,7 @@ struct mlx5_ib_mr {
 	struct mlx5_ib_dev     *dev;
 	u32 out[MLX5_ST_SZ_DW(create_mkey_out)];
 	struct mlx5_core_sig_ctx    *sig;
-	int			live;
+	unsigned int		live;
 	void			*descs_alloc;
 	int			access_flags; /* Needed for rereg MR */
 
@@ -639,7 +639,6 @@ struct mlx5_ib_mw {
 struct mlx5_ib_devx_mr {
 	struct mlx5_core_mkey	mmkey;
 	int			ndescs;
-	struct rcu_head		rcu;
 };
 
 struct mlx5_ib_umr_context {

drivers/infiniband/hw/mlx5/mr.c

@@ -84,32 +84,6 @@ static bool use_umr_mtt_update(struct mlx5_ib_mr *mr, u64 start, u64 length)
 		length + (start & (MLX5_ADAPTER_PAGE_SIZE - 1));
 }
 
-static void update_odp_mr(struct mlx5_ib_mr *mr)
-{
-	if (is_odp_mr(mr)) {
-		/*
-		 * This barrier prevents the compiler from moving the
-		 * setting of umem->odp_data->private to point to our
-		 * MR, before reg_umr finished, to ensure that the MR
-		 * initialization have finished before starting to
-		 * handle invalidations.
-		 */
-		smp_wmb();
-		to_ib_umem_odp(mr->umem)->private = mr;
-		/*
-		 * Make sure we will see the new
-		 * umem->odp_data->private value in the invalidation
-		 * routines, before we can get page faults on the
-		 * MR. Page faults can happen once we put the MR in
-		 * the tree, below this line. Without the barrier,
-		 * there can be a fault handling and an invalidation
-		 * before umem->odp_data->private == mr is visible to
-		 * the invalidation handler.
-		 */
-		smp_wmb();
-	}
-}
-
 static void reg_mr_callback(int status, struct mlx5_async_work *context)
 {
 	struct mlx5_ib_mr *mr =
@@ -1346,8 +1320,6 @@ struct ib_mr *mlx5_ib_reg_user_mr(struct ib_pd *pd, u64 start, u64 length,
 	mr->umem = umem;
 	set_mr_fields(dev, mr, npages, length, access_flags);
 
-	update_odp_mr(mr);
-
 	if (use_umr) {
 		int update_xlt_flags = MLX5_IB_UPD_XLT_ENABLE;
 
@@ -1363,10 +1335,12 @@ struct ib_mr *mlx5_ib_reg_user_mr(struct ib_pd *pd, u64 start, u64 length,
 		}
 	}
 
-	if (IS_ENABLED(CONFIG_INFINIBAND_ON_DEMAND_PAGING)) {
-		mr->live = 1;
+	if (is_odp_mr(mr)) {
+		to_ib_umem_odp(mr->umem)->private = mr;
 		atomic_set(&mr->num_pending_prefetch, 0);
 	}
+	if (IS_ENABLED(CONFIG_INFINIBAND_ON_DEMAND_PAGING))
+		smp_store_release(&mr->live, 1);
 
 	return &mr->ibmr;
 error:
@@ -1441,6 +1415,9 @@ int mlx5_ib_rereg_user_mr(struct ib_mr *ib_mr, int flags, u64 start,
 	if (!mr->umem)
 		return -EINVAL;
 
+	if (is_odp_mr(mr))
+		return -EOPNOTSUPP;
+
 	if (flags & IB_MR_REREG_TRANS) {
 		addr = virt_addr;
 		len = length;
@@ -1486,8 +1463,6 @@ int mlx5_ib_rereg_user_mr(struct ib_mr *ib_mr, int flags, u64 start,
 		}
 
 		mr->allocated_from_cache = 0;
-		if (IS_ENABLED(CONFIG_INFINIBAND_ON_DEMAND_PAGING))
-			mr->live = 1;
 	} else {
 		/*
 		 * Send a UMR WQE
@@ -1516,7 +1491,6 @@ int mlx5_ib_rereg_user_mr(struct ib_mr *ib_mr, int flags, u64 start,
 
 	set_mr_fields(dev, mr, npages, len, access_flags);
 
-	update_odp_mr(mr);
 	return 0;
 
 err:
@@ -1607,15 +1581,16 @@ static void dereg_mr(struct mlx5_ib_dev *dev, struct mlx5_ib_mr *mr)
 		/* Prevent new page faults and
 		 * prefetch requests from succeeding
 		 */
-		mr->live = 0;
+		WRITE_ONCE(mr->live, 0);
+
+		/* Wait for all running page-fault handlers to finish. */
+		synchronize_srcu(&dev->mr_srcu);
 
 		/* dequeue pending prefetch requests for the mr */
 		if (atomic_read(&mr->num_pending_prefetch))
 			flush_workqueue(system_unbound_wq);
 		WARN_ON(atomic_read(&mr->num_pending_prefetch));
 
-		/* Wait for all running page-fault handlers to finish. */
-		synchronize_srcu(&dev->mr_srcu);
 		/* Destroy all page mappings */
 		if (!umem_odp->is_implicit_odp)
 			mlx5_ib_invalidate_range(umem_odp,
@@ -1987,14 +1962,25 @@ struct ib_mw *mlx5_ib_alloc_mw(struct ib_pd *pd, enum ib_mw_type type,
 
 int mlx5_ib_dealloc_mw(struct ib_mw *mw)
 {
+	struct mlx5_ib_dev *dev = to_mdev(mw->device);
 	struct mlx5_ib_mw *mmw = to_mmw(mw);
 	int err;
 
-	err =  mlx5_core_destroy_mkey((to_mdev(mw->device))->mdev,
-				      &mmw->mmkey);
-	if (!err)
-		kfree(mmw);
-	return err;
+	if (IS_ENABLED(CONFIG_INFINIBAND_ON_DEMAND_PAGING)) {
+		xa_erase(&dev->mdev->priv.mkey_table,
+			 mlx5_base_mkey(mmw->mmkey.key));
+		/*
+		 * pagefault_single_data_segment() may be accessing mmw under
+		 * SRCU if the user bound an ODP MR to this MW.
+		 */
+		synchronize_srcu(&dev->mr_srcu);
+	}
+
+	err = mlx5_core_destroy_mkey(dev->mdev, &mmw->mmkey);
+	if (err)
+		return err;
+	kfree(mmw);
+	return 0;
 }
 
 int mlx5_ib_check_mr_status(struct ib_mr *ibmr, u32 check_mask,

drivers/infiniband/hw/mlx5/odp.c

@@ -178,6 +178,29 @@ void mlx5_odp_populate_klm(struct mlx5_klm *pklm, size_t offset,
 		return;
 	}
 
+	/*
+	 * The locking here is pretty subtle. Ideally the implicit children
+	 * list would be protected by the umem_mutex, however that is not
+	 * possible. Instead this uses a weaker update-then-lock pattern:
+	 *
+	 *  srcu_read_lock()
+	 *    <change children list>
+	 *    mutex_lock(umem_mutex)
+	 *     mlx5_ib_update_xlt()
+	 *    mutex_unlock(umem_mutex)
+	 *    destroy lkey
+	 *
+	 * ie any change the children list must be followed by the locked
+	 * update_xlt before destroying.
+	 *
+	 * The umem_mutex provides the acquire/release semantic needed to make
+	 * the children list visible to a racing thread. While SRCU is not
+	 * technically required, using it gives consistent use of the SRCU
+	 * locking around the children list.
+	 */
+	lockdep_assert_held(&to_ib_umem_odp(mr->umem)->umem_mutex);
+	lockdep_assert_held(&mr->dev->mr_srcu);
+
 	odp = odp_lookup(offset * MLX5_IMR_MTT_SIZE,
 			 nentries * MLX5_IMR_MTT_SIZE, mr);
 
@@ -202,15 +225,22 @@ static void mr_leaf_free_action(struct work_struct *work)
 	struct ib_umem_odp *odp = container_of(work, struct ib_umem_odp, work);
 	int idx = ib_umem_start(odp) >> MLX5_IMR_MTT_SHIFT;
 	struct mlx5_ib_mr *mr = odp->private, *imr = mr->parent;
+	struct ib_umem_odp *odp_imr = to_ib_umem_odp(imr->umem);
+	int srcu_key;
 
 	mr->parent = NULL;
 	synchronize_srcu(&mr->dev->mr_srcu);
 
-	ib_umem_odp_release(odp);
-	if (imr->live)
+	if (smp_load_acquire(&imr->live)) {
+		srcu_key = srcu_read_lock(&mr->dev->mr_srcu);
+		mutex_lock(&odp_imr->umem_mutex);
 		mlx5_ib_update_xlt(imr, idx, 1, 0,
 				   MLX5_IB_UPD_XLT_INDIRECT |
 				   MLX5_IB_UPD_XLT_ATOMIC);
+		mutex_unlock(&odp_imr->umem_mutex);
+		srcu_read_unlock(&mr->dev->mr_srcu, srcu_key);
+	}
+	ib_umem_odp_release(odp);
 	mlx5_mr_cache_free(mr->dev, mr);
 
 	if (atomic_dec_and_test(&imr->num_leaf_free))
@@ -278,7 +308,6 @@ void mlx5_ib_invalidate_range(struct ib_umem_odp *umem_odp, unsigned long start,
 				   idx - blk_start_idx + 1, 0,
 				   MLX5_IB_UPD_XLT_ZAP |
 				   MLX5_IB_UPD_XLT_ATOMIC);
-	mutex_unlock(&umem_odp->umem_mutex);
 	/*
 	 * We are now sure that the device will not access the
 	 * memory. We can safely unmap it, and mark it as dirty if
@@ -289,10 +318,12 @@ void mlx5_ib_invalidate_range(struct ib_umem_odp *umem_odp, unsigned long start,
 
 	if (unlikely(!umem_odp->npages && mr->parent &&
 		     !umem_odp->dying)) {
-		WRITE_ONCE(umem_odp->dying, 1);
+		WRITE_ONCE(mr->live, 0);
+		umem_odp->dying = 1;
 		atomic_inc(&mr->parent->num_leaf_free);
 		schedule_work(&umem_odp->work);
 	}
+	mutex_unlock(&umem_odp->umem_mutex);
 }
 
 void mlx5_ib_internal_fill_odp_caps(struct mlx5_ib_dev *dev)
@@ -429,8 +460,6 @@ static struct mlx5_ib_mr *implicit_mr_alloc(struct ib_pd *pd,
 	mr->ibmr.lkey = mr->mmkey.key;
 	mr->ibmr.rkey = mr->mmkey.key;
 
-	mr->live = 1;
-
 	mlx5_ib_dbg(dev, "key %x dev %p mr %p\n",
 		    mr->mmkey.key, dev->mdev, mr);
 
@@ -484,6 +513,8 @@ static struct ib_umem_odp *implicit_mr_get_data(struct mlx5_ib_mr *mr,
 		mtt->parent = mr;
 		INIT_WORK(&odp->work, mr_leaf_free_action);
 
+		smp_store_release(&mtt->live, 1);
+
 		if (!nentries)
 			start_idx = addr >> MLX5_IMR_MTT_SHIFT;
 		nentries++;
@@ -536,6 +567,7 @@ struct mlx5_ib_mr *mlx5_ib_alloc_implicit_mr(struct mlx5_ib_pd *pd,
 	init_waitqueue_head(&imr->q_leaf_free);
 	atomic_set(&imr->num_leaf_free, 0);
 	atomic_set(&imr->num_pending_prefetch, 0);
+	smp_store_release(&imr->live, 1);
 
 	return imr;
 }
@@ -555,15 +587,19 @@ void mlx5_ib_free_implicit_mr(struct mlx5_ib_mr *imr)
 		if (mr->parent != imr)
 			continue;
 
+		mutex_lock(&umem_odp->umem_mutex);
 		ib_umem_odp_unmap_dma_pages(umem_odp, ib_umem_start(umem_odp),
 					    ib_umem_end(umem_odp));
 
-		if (umem_odp->dying)
+		if (umem_odp->dying) {
+			mutex_unlock(&umem_odp->umem_mutex);
 			continue;
+		}
 
-		WRITE_ONCE(umem_odp->dying, 1);
+		umem_odp->dying = 1;
 		atomic_inc(&imr->num_leaf_free);
 		schedule_work(&umem_odp->work);
+		mutex_unlock(&umem_odp->umem_mutex);
 	}
 	up_read(&per_mm->umem_rwsem);
 
@@ -773,7 +809,7 @@ static int pagefault_single_data_segment(struct mlx5_ib_dev *dev,
 	switch (mmkey->type) {
 	case MLX5_MKEY_MR:
 		mr = container_of(mmkey, struct mlx5_ib_mr, mmkey);
-		if (!mr->live || !mr->ibmr.pd) {
+		if (!smp_load_acquire(&mr->live) || !mr->ibmr.pd) {
 			mlx5_ib_dbg(dev, "got dead MR\n");
 			ret = -EFAULT;
 			goto srcu_unlock;
@@ -1641,12 +1677,12 @@ static bool num_pending_prefetch_inc(struct ib_pd *pd,
 
 		mr = container_of(mmkey, struct mlx5_ib_mr, mmkey);
 
-		if (mr->ibmr.pd != pd) {
+		if (!smp_load_acquire(&mr->live)) {
 			ret = false;
 			break;
 		}
 
-		if (!mr->live) {
+		if (mr->ibmr.pd != pd) {
 			ret = false;
 			break;
 		}

drivers/infiniband/hw/vmw_pvrdma/pvrdma_srq.c

@@ -230,8 +230,6 @@ static void pvrdma_free_srq(struct pvrdma_dev *dev, struct pvrdma_srq *srq)
 
 	pvrdma_page_dir_cleanup(dev, &srq->pdir);
 
-	kfree(srq);
-
 	atomic_dec(&dev->num_srqs);
 }
 

drivers/infiniband/sw/siw/siw_qp.c

@@ -182,12 +182,19 @@ void siw_qp_llp_close(struct siw_qp *qp)
  */
 void siw_qp_llp_write_space(struct sock *sk)
 {
-	struct siw_cep *cep = sk_to_cep(sk);
+	struct siw_cep *cep;
 
-	cep->sk_write_space(sk);
+	read_lock(&sk->sk_callback_lock);
+
+	cep  = sk_to_cep(sk);
+	if (cep) {
+		cep->sk_write_space(sk);
 
-	if (!test_bit(SOCK_NOSPACE, &sk->sk_socket->flags))
-		(void)siw_sq_start(cep->qp);
+		if (!test_bit(SOCK_NOSPACE, &sk->sk_socket->flags))
+			(void)siw_sq_start(cep->qp);
+	}
+
+	read_unlock(&sk->sk_callback_lock);
 }
 
 static int siw_qp_readq_init(struct siw_qp *qp, int irq_size, int orq_size)

drivers/net/ethernet/mellanox/mlx5/core/mr.c

@@ -112,17 +112,11 @@ int mlx5_core_destroy_mkey(struct mlx5_core_dev *dev,
 	u32 out[MLX5_ST_SZ_DW(destroy_mkey_out)] = {0};
 	u32 in[MLX5_ST_SZ_DW(destroy_mkey_in)]   = {0};
 	struct xarray *mkeys = &dev->priv.mkey_table;
-	struct mlx5_core_mkey *deleted_mkey;
 	unsigned long flags;
 
 	xa_lock_irqsave(mkeys, flags);
-	deleted_mkey = __xa_erase(mkeys, mlx5_base_mkey(mkey->key));
+	__xa_erase(mkeys, mlx5_base_mkey(mkey->key));
 	xa_unlock_irqrestore(mkeys, flags);
-	if (!deleted_mkey) {
-		mlx5_core_dbg(dev, "failed xarray delete of mkey 0x%x\n",
-			      mlx5_base_mkey(mkey->key));
-		return -ENOENT;
-	}
 
 	MLX5_SET(destroy_mkey_in, in, opcode, MLX5_CMD_OP_DESTROY_MKEY);
 	MLX5_SET(destroy_mkey_in, in, mkey_index, mlx5_mkey_to_idx(mkey->key));