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

Pull main rdma updates from Doug Ledford:
 "This is the main pull request for the rdma stack this release.  The
  code has been through 0day and I had it tagged for linux-next testing
  for a couple days.

  Summary:

   - updates to mlx5

   - updates to mlx4 (two conflicts, both minor and easily resolved)

   - updates to iw_cxgb4 (one conflict, not so obvious to resolve,
     proper resolution is to keep the code in cxgb4_main.c as it is in
     Linus' tree as attach_uld was refactored and moved into
     cxgb4_uld.c)

   - improvements to uAPI (moved vendor specific API elements to uAPI
     area)

   - add hns-roce driver and hns and hns-roce ACPI reset support

   - conversion of all rdma code away from deprecated
     create_singlethread_workqueue

   - security improvement: remove unsafe ib_get_dma_mr (breaks lustre in
     staging)"

* tag 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/dledford/rdma: (75 commits)
  staging/lustre: Disable InfiniBand support
  iw_cxgb4: add fast-path for small REG_MR operations
  cxgb4: advertise support for FR_NSMR_TPTE_WR
  IB/core: correctly handle rdma_rw_init_mrs() failure
  IB/srp: Fix infinite loop when FMR sg[0].offset != 0
  IB/srp: Remove an unused argument
  IB/core: Improve ib_map_mr_sg() documentation
  IB/mlx4: Fix possible vl/sl field mismatch in LRH header in QP1 packets
  IB/mthca: Move user vendor structures
  IB/nes: Move user vendor structures
  IB/ocrdma: Move user vendor structures
  IB/mlx4: Move user vendor structures
  IB/cxgb4: Move user vendor structures
  IB/cxgb3: Move user vendor structures
  IB/mlx5: Move and decouple user vendor structures
  IB/{core,hw}: Add constant for node_desc
  ipoib: Make ipoib_warn ratelimited
  IB/mlx4/alias_GUID: Remove deprecated create_singlethread_workqueue
  IB/ipoib_verbs: Remove deprecated create_singlethread_workqueue
  IB/ipoib: Remove deprecated create_singlethread_workqueue
  ...

drivers/infiniband/ulp/ipoib/ipoib.h

@@ -772,7 +772,13 @@ static inline void ipoib_unregister_debugfs(void) { }
 #define ipoib_printk(level, priv, format, arg...)	\
 	printk(level "%s: " format, ((struct ipoib_dev_priv *) priv)->dev->name , ## arg)
 #define ipoib_warn(priv, format, arg...)		\
-	ipoib_printk(KERN_WARNING, priv, format , ## arg)
+do {							\
+	static DEFINE_RATELIMIT_STATE(_rs,		\
+		10 * HZ /*10 seconds */,		\
+		100);		\
+	if (__ratelimit(&_rs))				\
+		ipoib_printk(KERN_WARNING, priv, format , ## arg);\
+} while (0)
 
 extern int ipoib_sendq_size;
 extern int ipoib_recvq_size;

drivers/infiniband/ulp/ipoib/ipoib_main.c

@@ -2196,7 +2196,8 @@ static int __init ipoib_init_module(void)
 	 * its private workqueue, and we only queue up flush events
 	 * on our global flush workqueue.  This avoids the deadlocks.
 	 */
-	ipoib_workqueue = create_singlethread_workqueue("ipoib_flush");
+	ipoib_workqueue = alloc_ordered_workqueue("ipoib_flush",
+						  WQ_MEM_RECLAIM);
 	if (!ipoib_workqueue) {
 		ret = -ENOMEM;
 		goto err_fs;

drivers/infiniband/ulp/ipoib/ipoib_verbs.c

@@ -147,7 +147,7 @@ int ipoib_transport_dev_init(struct net_device *dev, struct ib_device *ca)
 	int ret, size;
 	int i;
 
-	priv->pd = ib_alloc_pd(priv->ca);
+	priv->pd = ib_alloc_pd(priv->ca, 0);
 	if (IS_ERR(priv->pd)) {
 		printk(KERN_WARNING "%s: failed to allocate PD\n", ca->name);
 		return -ENODEV;
@@ -157,7 +157,7 @@ int ipoib_transport_dev_init(struct net_device *dev, struct ib_device *ca)
 	 * the various IPoIB tasks assume they will never race against
 	 * themselves, so always use a single thread workqueue
 	 */
-	priv->wq = create_singlethread_workqueue("ipoib_wq");
+	priv->wq = alloc_ordered_workqueue("ipoib_wq", WQ_MEM_RECLAIM);
 	if (!priv->wq) {
 		printk(KERN_WARNING "ipoib: failed to allocate device WQ\n");
 		goto out_free_pd;

drivers/infiniband/ulp/iser/iscsi_iser.h

@@ -374,7 +374,6 @@ struct iser_reg_ops {
 struct iser_device {
 	struct ib_device             *ib_device;
 	struct ib_pd	             *pd;
-	struct ib_mr	             *mr;
 	struct ib_event_handler      event_handler;
 	struct list_head             ig_list;
 	int                          refcount;

drivers/infiniband/ulp/iser/iser_memory.c

@@ -199,7 +199,11 @@ iser_reg_dma(struct iser_device *device, struct iser_data_buf *mem,
 	 * FIXME: rework the registration code path to differentiate
 	 * rkey/lkey use cases
 	 */
-	reg->rkey = device->mr ? device->mr->rkey : 0;
+
+	if (device->pd->flags & IB_PD_UNSAFE_GLOBAL_RKEY)
+		reg->rkey = device->pd->unsafe_global_rkey;
+	else
+		reg->rkey = 0;
 	reg->sge.addr = ib_sg_dma_address(device->ib_device, &sg[0]);
 	reg->sge.length = ib_sg_dma_len(device->ib_device, &sg[0]);
 

drivers/infiniband/ulp/iser/iser_verbs.c

@@ -88,7 +88,8 @@ static int iser_create_device_ib_res(struct iser_device *device)
 		  device->comps_used, ib_dev->name,
 		  ib_dev->num_comp_vectors, max_cqe);
 
-	device->pd = ib_alloc_pd(ib_dev);
+	device->pd = ib_alloc_pd(ib_dev,
+		iser_always_reg ? 0 : IB_PD_UNSAFE_GLOBAL_RKEY);
 	if (IS_ERR(device->pd))
 		goto pd_err;
 
@@ -103,26 +104,13 @@ static int iser_create_device_ib_res(struct iser_device *device)
 		}
 	}
 
-	if (!iser_always_reg) {
-		int access = IB_ACCESS_LOCAL_WRITE |
-			     IB_ACCESS_REMOTE_WRITE |
-			     IB_ACCESS_REMOTE_READ;
-
-		device->mr = ib_get_dma_mr(device->pd, access);
-		if (IS_ERR(device->mr))
-			goto cq_err;
-	}
-
 	INIT_IB_EVENT_HANDLER(&device->event_handler, ib_dev,
 			      iser_event_handler);
 	if (ib_register_event_handler(&device->event_handler))
-		goto handler_err;
+		goto cq_err;
 
 	return 0;
 
-handler_err:
-	if (device->mr)
-		ib_dereg_mr(device->mr);
 cq_err:
 	for (i = 0; i < device->comps_used; i++) {
 		struct iser_comp *comp = &device->comps[i];
@@ -154,14 +142,10 @@ static void iser_free_device_ib_res(struct iser_device *device)
 	}
 
 	(void)ib_unregister_event_handler(&device->event_handler);
-	if (device->mr)
-		(void)ib_dereg_mr(device->mr);
 	ib_dealloc_pd(device->pd);
 
 	kfree(device->comps);
 	device->comps = NULL;
-
-	device->mr = NULL;
 	device->pd = NULL;
 }
 

drivers/infiniband/ulp/isert/ib_isert.c

@@ -309,7 +309,7 @@ isert_create_device_ib_res(struct isert_device *device)
 	if (ret)
 		goto out;
 
-	device->pd = ib_alloc_pd(ib_dev);
+	device->pd = ib_alloc_pd(ib_dev, 0);
 	if (IS_ERR(device->pd)) {
 		ret = PTR_ERR(device->pd);
 		isert_err("failed to allocate pd, device %p, ret=%d\n",

drivers/infiniband/ulp/srp/ib_srp.c

@@ -1262,6 +1262,7 @@ static int srp_map_finish_fmr(struct srp_map_state *state,
 {
 	struct srp_target_port *target = ch->target;
 	struct srp_device *dev = target->srp_host->srp_dev;
+	struct ib_pd *pd = target->pd;
 	struct ib_pool_fmr *fmr;
 	u64 io_addr = 0;
 
@@ -1273,9 +1274,9 @@ static int srp_map_finish_fmr(struct srp_map_state *state,
 	if (state->npages == 0)
 		return 0;
 
-	if (state->npages == 1 && target->global_mr) {
+	if (state->npages == 1 && (pd->flags & IB_PD_UNSAFE_GLOBAL_RKEY)) {
 		srp_map_desc(state, state->base_dma_addr, state->dma_len,
-			     target->global_mr->rkey);
+			     pd->unsafe_global_rkey);
 		goto reset_state;
 	}
 
@@ -1315,6 +1316,7 @@ static int srp_map_finish_fr(struct srp_map_state *state,
 {
 	struct srp_target_port *target = ch->target;
 	struct srp_device *dev = target->srp_host->srp_dev;
+	struct ib_pd *pd = target->pd;
 	struct ib_send_wr *bad_wr;
 	struct ib_reg_wr wr;
 	struct srp_fr_desc *desc;
@@ -1326,12 +1328,12 @@ static int srp_map_finish_fr(struct srp_map_state *state,
 
 	WARN_ON_ONCE(!dev->use_fast_reg);
 
-	if (sg_nents == 1 && target->global_mr) {
+	if (sg_nents == 1 && (pd->flags & IB_PD_UNSAFE_GLOBAL_RKEY)) {
 		unsigned int sg_offset = sg_offset_p ? *sg_offset_p : 0;
 
 		srp_map_desc(state, sg_dma_address(state->sg) + sg_offset,
 			     sg_dma_len(state->sg) - sg_offset,
-			     target->global_mr->rkey);
+			     pd->unsafe_global_rkey);
 		if (sg_offset_p)
 			*sg_offset_p = 0;
 		return 1;
@@ -1386,7 +1388,7 @@ static int srp_map_finish_fr(struct srp_map_state *state,
 
 static int srp_map_sg_entry(struct srp_map_state *state,
 			    struct srp_rdma_ch *ch,
-			    struct scatterlist *sg, int sg_index)
+			    struct scatterlist *sg)
 {
 	struct srp_target_port *target = ch->target;
 	struct srp_device *dev = target->srp_host->srp_dev;
@@ -1400,7 +1402,9 @@ static int srp_map_sg_entry(struct srp_map_state *state,
 
 	while (dma_len) {
 		unsigned offset = dma_addr & ~dev->mr_page_mask;
-		if (state->npages == dev->max_pages_per_mr || offset != 0) {
+
+		if (state->npages == dev->max_pages_per_mr ||
+		    (state->npages > 0 && offset != 0)) {
 			ret = srp_map_finish_fmr(state, ch);
 			if (ret)
 				return ret;
@@ -1417,12 +1421,12 @@ static int srp_map_sg_entry(struct srp_map_state *state,
 	}
 
 	/*
-	 * If the last entry of the MR wasn't a full page, then we need to
+	 * If the end of the MR is not on a page boundary then we need to
 	 * close it out and start a new one -- we can only merge at page
 	 * boundaries.
 	 */
 	ret = 0;
-	if (len != dev->mr_page_size)
+	if ((dma_addr & ~dev->mr_page_mask) != 0)
 		ret = srp_map_finish_fmr(state, ch);
 	return ret;
 }
@@ -1439,7 +1443,7 @@ static int srp_map_sg_fmr(struct srp_map_state *state, struct srp_rdma_ch *ch,
 	state->fmr.end = req->fmr_list + ch->target->mr_per_cmd;
 
 	for_each_sg(scat, sg, count, i) {
-		ret = srp_map_sg_entry(state, ch, sg, i);
+		ret = srp_map_sg_entry(state, ch, sg);
 		if (ret)
 			return ret;
 	}
@@ -1491,7 +1495,7 @@ static int srp_map_sg_dma(struct srp_map_state *state, struct srp_rdma_ch *ch,
 	for_each_sg(scat, sg, count, i) {
 		srp_map_desc(state, ib_sg_dma_address(dev->dev, sg),
 			     ib_sg_dma_len(dev->dev, sg),
-			     target->global_mr->rkey);
+			     target->pd->unsafe_global_rkey);
 	}
 
 	return 0;
@@ -1591,6 +1595,7 @@ static int srp_map_data(struct scsi_cmnd *scmnd, struct srp_rdma_ch *ch,
 			struct srp_request *req)
 {
 	struct srp_target_port *target = ch->target;
+	struct ib_pd *pd = target->pd;
 	struct scatterlist *scat;
 	struct srp_cmd *cmd = req->cmd->buf;
 	int len, nents, count, ret;
@@ -1626,7 +1631,7 @@ static int srp_map_data(struct scsi_cmnd *scmnd, struct srp_rdma_ch *ch,
 	fmt = SRP_DATA_DESC_DIRECT;
 	len = sizeof (struct srp_cmd) +	sizeof (struct srp_direct_buf);
 
-	if (count == 1 && target->global_mr) {
+	if (count == 1 && (pd->flags & IB_PD_UNSAFE_GLOBAL_RKEY)) {
 		/*
 		 * The midlayer only generated a single gather/scatter
 		 * entry, or DMA mapping coalesced everything to a
@@ -1636,7 +1641,7 @@ static int srp_map_data(struct scsi_cmnd *scmnd, struct srp_rdma_ch *ch,
 		struct srp_direct_buf *buf = (void *) cmd->add_data;
 
 		buf->va  = cpu_to_be64(ib_sg_dma_address(ibdev, scat));
-		buf->key = cpu_to_be32(target->global_mr->rkey);
+		buf->key = cpu_to_be32(pd->unsafe_global_rkey);
 		buf->len = cpu_to_be32(ib_sg_dma_len(ibdev, scat));
 
 		req->nmdesc = 0;
@@ -1709,14 +1714,14 @@ static int srp_map_data(struct scsi_cmnd *scmnd, struct srp_rdma_ch *ch,
 	memcpy(indirect_hdr->desc_list, req->indirect_desc,
 	       count * sizeof (struct srp_direct_buf));
 
-	if (!target->global_mr) {
+	if (!(pd->flags & IB_PD_UNSAFE_GLOBAL_RKEY)) {
 		ret = srp_map_idb(ch, req, state.gen.next, state.gen.end,
 				  idb_len, &idb_rkey);
 		if (ret < 0)
 			goto unmap;
 		req->nmdesc++;
 	} else {
-		idb_rkey = cpu_to_be32(target->global_mr->rkey);
+		idb_rkey = cpu_to_be32(pd->unsafe_global_rkey);
 	}
 
 	indirect_hdr->table_desc.va = cpu_to_be64(req->indirect_dma_addr);
@@ -3268,8 +3273,8 @@ static ssize_t srp_create_target(struct device *dev,
 	target->io_class	= SRP_REV16A_IB_IO_CLASS;
 	target->scsi_host	= target_host;
 	target->srp_host	= host;
+	target->pd		= host->srp_dev->pd;
 	target->lkey		= host->srp_dev->pd->local_dma_lkey;
-	target->global_mr	= host->srp_dev->global_mr;
 	target->cmd_sg_cnt	= cmd_sg_entries;
 	target->sg_tablesize	= indirect_sg_entries ? : cmd_sg_entries;
 	target->allow_ext_sg	= allow_ext_sg;
@@ -3524,6 +3529,7 @@ static void srp_add_one(struct ib_device *device)
 	struct srp_host *host;
 	int mr_page_shift, p;
 	u64 max_pages_per_mr;
+	unsigned int flags = 0;
 
 	srp_dev = kzalloc(sizeof(*srp_dev), GFP_KERNEL);
 	if (!srp_dev)
@@ -3558,6 +3564,10 @@ static void srp_add_one(struct ib_device *device)
 		srp_dev->use_fmr = !srp_dev->use_fast_reg && srp_dev->has_fmr;
 	}
 
+	if (never_register || !register_always ||
+	    (!srp_dev->has_fmr && !srp_dev->has_fr))
+		flags |= IB_PD_UNSAFE_GLOBAL_RKEY;
+
 	if (srp_dev->use_fast_reg) {
 		srp_dev->max_pages_per_mr =
 			min_t(u32, srp_dev->max_pages_per_mr,
@@ -3573,19 +3583,10 @@ static void srp_add_one(struct ib_device *device)
 	INIT_LIST_HEAD(&srp_dev->dev_list);
 
 	srp_dev->dev = device;
-	srp_dev->pd  = ib_alloc_pd(device);
+	srp_dev->pd  = ib_alloc_pd(device, flags);
 	if (IS_ERR(srp_dev->pd))
 		goto free_dev;
 
-	if (never_register || !register_always ||
-	    (!srp_dev->has_fmr && !srp_dev->has_fr)) {
-		srp_dev->global_mr = ib_get_dma_mr(srp_dev->pd,
-						   IB_ACCESS_LOCAL_WRITE |
-						   IB_ACCESS_REMOTE_READ |
-						   IB_ACCESS_REMOTE_WRITE);
-		if (IS_ERR(srp_dev->global_mr))
-			goto err_pd;
-	}
 
 	for (p = rdma_start_port(device); p <= rdma_end_port(device); ++p) {
 		host = srp_add_port(srp_dev, p);
@@ -3596,9 +3597,6 @@ static void srp_add_one(struct ib_device *device)
 	ib_set_client_data(device, &srp_client, srp_dev);
 	return;
 
-err_pd:
-	ib_dealloc_pd(srp_dev->pd);
-
 free_dev:
 	kfree(srp_dev);
 }
@@ -3638,8 +3636,6 @@ static void srp_remove_one(struct ib_device *device, void *client_data)
 		kfree(host);
 	}
 
-	if (srp_dev->global_mr)
-		ib_dereg_mr(srp_dev->global_mr);
 	ib_dealloc_pd(srp_dev->pd);
 
 	kfree(srp_dev);

drivers/infiniband/ulp/srp/ib_srp.h

@@ -90,7 +90,6 @@ struct srp_device {
 	struct list_head	dev_list;
 	struct ib_device       *dev;
 	struct ib_pd	       *pd;
-	struct ib_mr	       *global_mr;
 	u64			mr_page_mask;
 	int			mr_page_size;
 	int			mr_max_size;
@@ -179,7 +178,7 @@ struct srp_target_port {
 	spinlock_t		lock;
 
 	/* read only in the hot path */
-	struct ib_mr		*global_mr;
+	struct ib_pd		*pd;
 	struct srp_rdma_ch	*ch;
 	u32			ch_count;
 	u32			lkey;

drivers/infiniband/ulp/srpt/ib_srpt.c

@@ -2480,7 +2480,7 @@ static void srpt_add_one(struct ib_device *device)
 	init_waitqueue_head(&sdev->ch_releaseQ);
 	mutex_init(&sdev->mutex);
 
-	sdev->pd = ib_alloc_pd(device);
+	sdev->pd = ib_alloc_pd(device, 0);
 	if (IS_ERR(sdev->pd))
 		goto free_dev;