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

Pull rdma updates from Jason Gunthorpe:
 "This has been a smaller cycle than normal. One new driver was
  accepted, which is unusual, and at least one more driver remains in
  review on the list.

  Summary:

   - Driver fixes for hns, hfi1, nes, rxe, i40iw, mlx5, cxgb4,
     vmw_pvrdma

   - Many patches from MatthewW converting radix tree and IDR users to
     use xarray

   - Introduction of tracepoints to the MAD layer

   - Build large SGLs at the start for DMA mapping and get the driver to
     split them

   - Generally clean SGL handling code throughout the subsystem

   - Support for restricting RDMA devices to net namespaces for
     containers

   - Progress to remove object allocation boilerplate code from drivers

   - Change in how the mlx5 driver shows representor ports linked to VFs

   - mlx5 uapi feature to access the on chip SW ICM memory

   - Add a new driver for 'EFA'. This is HW that supports user space
     packet processing through QPs in Amazon's cloud"

* tag 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/rdma/rdma: (186 commits)
  RDMA/ipoib: Allow user space differentiate between valid dev_port
  IB/core, ipoib: Do not overreact to SM LID change event
  RDMA/device: Don't fire uevent before device is fully initialized
  lib/scatterlist: Remove leftover from sg_page_iter comment
  RDMA/efa: Add driver to Kconfig/Makefile
  RDMA/efa: Add the efa module
  RDMA/efa: Add EFA verbs implementation
  RDMA/efa: Add common command handlers
  RDMA/efa: Implement functions that submit and complete admin commands
  RDMA/efa: Add the ABI definitions
  RDMA/efa: Add the com service API definitions
  RDMA/efa: Add the efa_com.h file
  RDMA/efa: Add the efa.h header file
  RDMA/efa: Add EFA device definitions
  RDMA: Add EFA related definitions
  RDMA/umem: Remove hugetlb flag
  RDMA/bnxt_re: Use core helpers to get aligned DMA address
  RDMA/i40iw: Use core helpers to get aligned DMA address within a supported page size
  RDMA/verbs: Add a DMA iterator to return aligned contiguous memory blocks
  RDMA/umem: Add API to find best driver supported page size in an MR
  ...

5 files changed, 64 insertions, 90 deletions

diff --git a/drivers/infiniband/hw/i40iw/i40iw.h b/drivers/infiniband/hw/i40iw/i40iw.h
index 2f2b4426ded7..8feec35f95a7 100644
--- a/drivers/infiniband/hw/i40iw/i40iw.h
+++ b/drivers/infiniband/hw/i40iw/i40iw.h
@@ -552,7 +552,7 @@ enum i40iw_status_code i40iw_obj_aligned_mem(struct i40iw_device *iwdev,
 
 void i40iw_request_reset(struct i40iw_device *iwdev);
 void i40iw_destroy_rdma_device(struct i40iw_ib_device *iwibdev);
-void i40iw_setup_cm_core(struct i40iw_device *iwdev);
+int i40iw_setup_cm_core(struct i40iw_device *iwdev);
 void i40iw_cleanup_cm_core(struct i40iw_cm_core *cm_core);
 void i40iw_process_ceq(struct i40iw_device *, struct i40iw_ceq *iwceq);
 void i40iw_process_aeq(struct i40iw_device *);
diff --git a/drivers/infiniband/hw/i40iw/i40iw_cm.c b/drivers/infiniband/hw/i40iw/i40iw_cm.c
index 206cfb0016f8..8233f5a4e623 100644
--- a/drivers/infiniband/hw/i40iw/i40iw_cm.c
+++ b/drivers/infiniband/hw/i40iw/i40iw_cm.c
@@ -3237,7 +3237,7 @@ void i40iw_receive_ilq(struct i40iw_sc_vsi *vsi, struct i40iw_puda_buf *rbuf)
  * core
  * @iwdev: iwarp device structure
  */
-void i40iw_setup_cm_core(struct i40iw_device *iwdev)
+int i40iw_setup_cm_core(struct i40iw_device *iwdev)
 {
 	struct i40iw_cm_core *cm_core = &iwdev->cm_core;
 
@@ -3256,9 +3256,19 @@ void i40iw_setup_cm_core(struct i40iw_device *iwdev)
 
 	cm_core->event_wq = alloc_ordered_workqueue("iwewq",
 						    WQ_MEM_RECLAIM);
+	if (!cm_core->event_wq)
+		goto error;
 
 	cm_core->disconn_wq = alloc_ordered_workqueue("iwdwq",
 						      WQ_MEM_RECLAIM);
+	if (!cm_core->disconn_wq)
+		goto error;
+
+	return 0;
+error:
+	i40iw_cleanup_cm_core(&iwdev->cm_core);
+
+	return -ENOMEM;
 }
 
 /**
@@ -3278,8 +3288,10 @@ void i40iw_cleanup_cm_core(struct i40iw_cm_core *cm_core)
 		del_timer_sync(&cm_core->tcp_timer);
 	spin_unlock_irqrestore(&cm_core->ht_lock, flags);
 
-	destroy_workqueue(cm_core->event_wq);
-	destroy_workqueue(cm_core->disconn_wq);
+	if (cm_core->event_wq)
+		destroy_workqueue(cm_core->event_wq);
+	if (cm_core->disconn_wq)
+		destroy_workqueue(cm_core->disconn_wq);
 }
 
 /**
@@ -3478,7 +3490,8 @@ static void i40iw_qp_disconnect(struct i40iw_qp *iwqp)
 		/* Need to free the Last Streaming Mode Message */
 		if (iwqp->ietf_mem.va) {
 			if (iwqp->lsmm_mr)
-				iwibdev->ibdev.ops.dereg_mr(iwqp->lsmm_mr);
+				iwibdev->ibdev.ops.dereg_mr(iwqp->lsmm_mr,
+							    NULL);
 			i40iw_free_dma_mem(iwdev->sc_dev.hw, &iwqp->ietf_mem);
 		}
 	}
diff --git a/drivers/infiniband/hw/i40iw/i40iw_main.c b/drivers/infiniband/hw/i40iw/i40iw_main.c
index 68095f00d08f..10932baee279 100644
--- a/drivers/infiniband/hw/i40iw/i40iw_main.c
+++ b/drivers/infiniband/hw/i40iw/i40iw_main.c
@@ -1641,7 +1641,10 @@ static int i40iw_open(struct i40e_info *ldev, struct i40e_client *client)
 	iwdev = &hdl->device;
 	iwdev->hdl = hdl;
 	dev = &iwdev->sc_dev;
-	i40iw_setup_cm_core(iwdev);
+	if (i40iw_setup_cm_core(iwdev)) {
+		kfree(iwdev->hdl);
+		return -ENOMEM;
+	}
 
 	dev->back_dev = (void *)iwdev;
 	iwdev->ldev = &hdl->ldev;
diff --git a/drivers/infiniband/hw/i40iw/i40iw_verbs.c b/drivers/infiniband/hw/i40iw/i40iw_verbs.c
index a8352e3ca23d..5689d742bafb 100644
--- a/drivers/infiniband/hw/i40iw/i40iw_verbs.c
+++ b/drivers/infiniband/hw/i40iw/i40iw_verbs.c
@@ -291,18 +291,15 @@ static void i40iw_dealloc_push_page(struct i40iw_device *iwdev, struct i40iw_sc_
 /**
  * i40iw_alloc_pd - allocate protection domain
  * @pd: PD pointer
- * @context: user context created during alloc
  * @udata: user data
  */
-static int i40iw_alloc_pd(struct ib_pd *pd, struct ib_ucontext *context,
-			  struct ib_udata *udata)
+static int i40iw_alloc_pd(struct ib_pd *pd, struct ib_udata *udata)
 {
 	struct i40iw_pd *iwpd = to_iwpd(pd);
 	struct i40iw_device *iwdev = to_iwdev(pd->device);
 	struct i40iw_sc_dev *dev = &iwdev->sc_dev;
 	struct i40iw_alloc_pd_resp uresp;
 	struct i40iw_sc_pd *sc_pd;
-	struct i40iw_ucontext *ucontext;
 	u32 pd_id = 0;
 	int err;
 
@@ -318,8 +315,9 @@ static int i40iw_alloc_pd(struct ib_pd *pd, struct ib_ucontext *context,
 
 	sc_pd = &iwpd->sc_pd;
 
-	if (context) {
-		ucontext = to_ucontext(context);
+	if (udata) {
+		struct i40iw_ucontext *ucontext = rdma_udata_to_drv_context(
+			udata, struct i40iw_ucontext, ibucontext);
 		dev->iw_pd_ops->pd_init(dev, sc_pd, pd_id, ucontext->abi_ver);
 		memset(&uresp, 0, sizeof(uresp));
 		uresp.pd_id = pd_id;
@@ -342,8 +340,9 @@ error:
 /**
  * i40iw_dealloc_pd - deallocate pd
  * @ibpd: ptr of pd to be deallocated
+ * @udata: user data or null for kernel object
  */
-static void i40iw_dealloc_pd(struct ib_pd *ibpd)
+static void i40iw_dealloc_pd(struct ib_pd *ibpd, struct ib_udata *udata)
 {
 	struct i40iw_pd *iwpd = to_iwpd(ibpd);
 	struct i40iw_device *iwdev = to_iwdev(ibpd->device);
@@ -413,7 +412,7 @@ static void i40iw_clean_cqes(struct i40iw_qp *iwqp, struct i40iw_cq *iwcq)
  * i40iw_destroy_qp - destroy qp
  * @ibqp: qp's ib pointer also to get to device's qp address
  */
-static int i40iw_destroy_qp(struct ib_qp *ibqp)
+static int i40iw_destroy_qp(struct ib_qp *ibqp, struct ib_udata *udata)
 {
 	struct i40iw_qp *iwqp = to_iwqp(ibqp);
 
@@ -744,8 +743,8 @@ static struct ib_qp *i40iw_create_qp(struct ib_pd *ibpd,
 		err_code = ib_copy_to_udata(udata, &uresp, sizeof(uresp));
 		if (err_code) {
 			i40iw_pr_err("copy_to_udata failed\n");
-			i40iw_destroy_qp(&iwqp->ibqp);
-			   /* let the completion of the qp destroy free the qp */
+			i40iw_destroy_qp(&iwqp->ibqp, udata);
+			/* let the completion of the qp destroy free the qp */
 			return ERR_PTR(err_code);
 		}
 	}
@@ -1063,8 +1062,9 @@ void i40iw_cq_wq_destroy(struct i40iw_device *iwdev, struct i40iw_sc_cq *cq)
 /**
  * i40iw_destroy_cq - destroy cq
  * @ib_cq: cq pointer
+ * @udata: user data or NULL for kernel object
  */
-static int i40iw_destroy_cq(struct ib_cq *ib_cq)
+static int i40iw_destroy_cq(struct ib_cq *ib_cq, struct ib_udata *udata)
 {
 	struct i40iw_cq *iwcq;
 	struct i40iw_device *iwdev;
@@ -1089,12 +1089,10 @@ static int i40iw_destroy_cq(struct ib_cq *ib_cq)
  * i40iw_create_cq - create cq
  * @ibdev: device pointer from stack
  * @attr: attributes for cq
- * @context: user context created during alloc
  * @udata: user data
  */
 static struct ib_cq *i40iw_create_cq(struct ib_device *ibdev,
 				     const struct ib_cq_init_attr *attr,
-				     struct ib_ucontext *context,
 				     struct ib_udata *udata)
 {
 	struct i40iw_device *iwdev = to_iwdev(ibdev);
@@ -1144,14 +1142,14 @@ static struct ib_cq *i40iw_create_cq(struct ib_device *ibdev,
 	info.ceq_id_valid = true;
 	info.ceqe_mask = 1;
 	info.type = I40IW_CQ_TYPE_IWARP;
-	if (context) {
-		struct i40iw_ucontext *ucontext;
+	if (udata) {
+		struct i40iw_ucontext *ucontext = rdma_udata_to_drv_context(
+			udata, struct i40iw_ucontext, ibucontext);
 		struct i40iw_create_cq_req req;
 		struct i40iw_cq_mr *cqmr;
 
 		memset(&req, 0, sizeof(req));
 		iwcq->user_mode = true;
-		ucontext = to_ucontext(context);
 		if (ib_copy_from_udata(&req, udata, sizeof(struct i40iw_create_cq_req))) {
 			err_code = -EFAULT;
 			goto cq_free_resources;
@@ -1221,7 +1219,7 @@ static struct ib_cq *i40iw_create_cq(struct ib_device *ibdev,
 		goto cq_free_resources;
 	}
 
-	if (context) {
+	if (udata) {
 		struct i40iw_create_cq_resp resp;
 
 		memset(&resp, 0, sizeof(resp));
@@ -1340,53 +1338,22 @@ static void i40iw_copy_user_pgaddrs(struct i40iw_mr *iwmr,
 	struct i40iw_pbl *iwpbl = &iwmr->iwpbl;
 	struct i40iw_pble_alloc *palloc = &iwpbl->pble_alloc;
 	struct i40iw_pble_info *pinfo;
-	struct sg_dma_page_iter sg_iter;
-	u64 pg_addr = 0;
+	struct ib_block_iter biter;
 	u32 idx = 0;
-	bool first_pg = true;
 
 	pinfo = (level == I40IW_LEVEL_1) ? NULL : palloc->level2.leaf;
 
 	if (iwmr->type == IW_MEMREG_TYPE_QP)
 		iwpbl->qp_mr.sq_page = sg_page(region->sg_head.sgl);
 
-	for_each_sg_dma_page (region->sg_head.sgl, &sg_iter, region->nmap, 0) {
-		pg_addr = sg_page_iter_dma_address(&sg_iter);
-		if (first_pg)
-			*pbl = cpu_to_le64(pg_addr & iwmr->page_msk);
-		else if (!(pg_addr & ~iwmr->page_msk))
-			*pbl = cpu_to_le64(pg_addr);
-		else
-			continue;
-
-		first_pg = false;
+	rdma_for_each_block(region->sg_head.sgl, &biter, region->nmap,
+			    iwmr->page_size) {
+		*pbl = rdma_block_iter_dma_address(&biter);
 		pbl = i40iw_next_pbl_addr(pbl, &pinfo, &idx);
 	}
 }
 
 /**
- * i40iw_set_hugetlb_params - set MR pg size and mask to huge pg values.
- * @addr: virtual address
- * @iwmr: mr pointer for this memory registration
- */
-static void i40iw_set_hugetlb_values(u64 addr, struct i40iw_mr *iwmr)
-{
-	struct vm_area_struct *vma;
-	struct hstate *h;
-
-	down_read(&current->mm->mmap_sem);
-	vma = find_vma(current->mm, addr);
-	if (vma && is_vm_hugetlb_page(vma)) {
-		h = hstate_vma(vma);
-		if (huge_page_size(h) == 0x200000) {
-			iwmr->page_size = huge_page_size(h);
-			iwmr->page_msk = huge_page_mask(h);
-		}
-	}
-	up_read(&current->mm->mmap_sem);
-}
-
-/**
  * i40iw_check_mem_contiguous - check if pbls stored in arr are contiguous
  * @arr: lvl1 pbl array
  * @npages: page count
@@ -1601,10 +1568,10 @@ static int i40iw_hw_alloc_stag(struct i40iw_device *iwdev, struct i40iw_mr *iwmr
  * @pd: ibpd pointer
  * @mr_type: memory for stag registrion
  * @max_num_sg: man number of pages
+ * @udata: user data or NULL for kernel objects
  */
-static struct ib_mr *i40iw_alloc_mr(struct ib_pd *pd,
-				    enum ib_mr_type mr_type,
-				    u32 max_num_sg)
+static struct ib_mr *i40iw_alloc_mr(struct ib_pd *pd, enum ib_mr_type mr_type,
+				    u32 max_num_sg, struct ib_udata *udata)
 {
 	struct i40iw_pd *iwpd = to_iwpd(pd);
 	struct i40iw_device *iwdev = to_iwdev(pd->device);
@@ -1841,10 +1808,9 @@ static struct ib_mr *i40iw_reg_user_mr(struct ib_pd *pd,
 	iwmr->ibmr.device = pd->device;
 
 	iwmr->page_size = PAGE_SIZE;
-	iwmr->page_msk = PAGE_MASK;
-
-	if (region->hugetlb && (req.reg_type == IW_MEMREG_TYPE_MEM))
-		i40iw_set_hugetlb_values(start, iwmr);
+	if (req.reg_type == IW_MEMREG_TYPE_MEM)
+		iwmr->page_size = ib_umem_find_best_pgsz(region, SZ_4K | SZ_2M,
+							 virt);
 
 	region_length = region->length + (start & (iwmr->page_size - 1));
 	pg_shift = ffs(iwmr->page_size) - 1;
@@ -2038,7 +2004,7 @@ static void i40iw_del_memlist(struct i40iw_mr *iwmr,
  * i40iw_dereg_mr - deregister mr
  * @ib_mr: mr ptr for dereg
  */
-static int i40iw_dereg_mr(struct ib_mr *ib_mr)
+static int i40iw_dereg_mr(struct ib_mr *ib_mr, struct ib_udata *udata)
 {
 	struct ib_pd *ibpd = ib_mr->pd;
 	struct i40iw_pd *iwpd = to_iwpd(ibpd);
@@ -2058,9 +2024,12 @@ static int i40iw_dereg_mr(struct ib_mr *ib_mr)
 	if (iwmr->type != IW_MEMREG_TYPE_MEM) {
 		/* region is released. only test for userness. */
 		if (iwmr->region) {
-			struct i40iw_ucontext *ucontext;
+			struct i40iw_ucontext *ucontext =
+				rdma_udata_to_drv_context(
+					udata,
+					struct i40iw_ucontext,
+					ibucontext);
 
-			ucontext = to_ucontext(ibpd->uobject->context);
 			i40iw_del_memlist(iwmr, ucontext);
 		}
 		if (iwpbl->pbl_allocated && iwmr->type != IW_MEMREG_TYPE_QP)
@@ -2703,6 +2672,14 @@ static const struct ib_device_ops i40iw_dev_ops = {
 	.get_dma_mr = i40iw_get_dma_mr,
 	.get_hw_stats = i40iw_get_hw_stats,
 	.get_port_immutable = i40iw_port_immutable,
+	.iw_accept = i40iw_accept,
+	.iw_add_ref = i40iw_add_ref,
+	.iw_connect = i40iw_connect,
+	.iw_create_listen = i40iw_create_listen,
+	.iw_destroy_listen = i40iw_destroy_listen,
+	.iw_get_qp = i40iw_get_qp,
+	.iw_reject = i40iw_reject,
+	.iw_rem_ref = i40iw_rem_ref,
 	.map_mr_sg = i40iw_map_mr_sg,
 	.mmap = i40iw_mmap,
 	.modify_qp = i40iw_modify_qp,
@@ -2766,22 +2743,8 @@ static struct i40iw_ib_device *i40iw_init_rdma_device(struct i40iw_device *iwdev
 	iwibdev->ibdev.phys_port_cnt = 1;
 	iwibdev->ibdev.num_comp_vectors = iwdev->ceqs_count;
 	iwibdev->ibdev.dev.parent = &pcidev->dev;
-	iwibdev->ibdev.iwcm = kzalloc(sizeof(*iwibdev->ibdev.iwcm), GFP_KERNEL);
-	if (!iwibdev->ibdev.iwcm) {
-		ib_dealloc_device(&iwibdev->ibdev);
-		return NULL;
-	}
-
-	iwibdev->ibdev.iwcm->add_ref = i40iw_add_ref;
-	iwibdev->ibdev.iwcm->rem_ref = i40iw_rem_ref;
-	iwibdev->ibdev.iwcm->get_qp = i40iw_get_qp;
-	iwibdev->ibdev.iwcm->connect = i40iw_connect;
-	iwibdev->ibdev.iwcm->accept = i40iw_accept;
-	iwibdev->ibdev.iwcm->reject = i40iw_reject;
-	iwibdev->ibdev.iwcm->create_listen = i40iw_create_listen;
-	iwibdev->ibdev.iwcm->destroy_listen = i40iw_destroy_listen;
-	memcpy(iwibdev->ibdev.iwcm->ifname, netdev->name,
-	       sizeof(iwibdev->ibdev.iwcm->ifname));
+	memcpy(iwibdev->ibdev.iw_ifname, netdev->name,
+	       sizeof(iwibdev->ibdev.iw_ifname));
 	ib_set_device_ops(&iwibdev->ibdev, &i40iw_dev_ops);
 
 	return iwibdev;
@@ -2812,8 +2775,6 @@ void i40iw_destroy_rdma_device(struct i40iw_ib_device *iwibdev)
 		return;
 
 	ib_unregister_device(&iwibdev->ibdev);
-	kfree(iwibdev->ibdev.iwcm);
-	iwibdev->ibdev.iwcm = NULL;
 	wait_event_timeout(iwibdev->iwdev->close_wq,
 			   !atomic64_read(&iwibdev->iwdev->use_count),
 			   I40IW_EVENT_TIMEOUT);
@@ -2841,8 +2802,6 @@ int i40iw_register_rdma_device(struct i40iw_device *iwdev)
 
 	return 0;
 error:
-	kfree(iwdev->iwibdev->ibdev.iwcm);
-	iwdev->iwibdev->ibdev.iwcm = NULL;
 	ib_dealloc_device(&iwdev->iwibdev->ibdev);
 	return ret;
 }
diff --git a/drivers/infiniband/hw/i40iw/i40iw_verbs.h b/drivers/infiniband/hw/i40iw/i40iw_verbs.h
index 76cf173377ab..3a413752ccc3 100644
--- a/drivers/infiniband/hw/i40iw/i40iw_verbs.h
+++ b/drivers/infiniband/hw/i40iw/i40iw_verbs.h
@@ -94,8 +94,7 @@ struct i40iw_mr {
 	struct ib_umem *region;
 	u16 type;
 	u32 page_cnt;
-	u32 page_size;
-	u64 page_msk;
+	u64 page_size;
 	u32 npages;
 	u32 stag;
 	u64 length;