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 with many of the commits being smallish
  code fixes and improvements across the drivers.

   - Driver updates for bnxt_re, cxgb4, hfi1, hns, mlx5, nes, qedr, and
     rxe

   - Memory window support in hns

   - mlx5 user API 'flow mutate/steering' allows accessing the full
     packet mangling and matching machinery from user space

   - Support inter-working with verbs API calls in the 'devx' mlx5 user
     API, and provide options to use devx with less privilege

   - Modernize the use of syfs and the device interface to use attribute
     groups and cdev properly for uverbs, and clean up some of the core
     code's device list management

   - More progress on net namespaces for RDMA devices

   - Consolidate driver BAR mmapping support into core code helpers and
     rework how RDMA holds poitners to mm_struct for get_user_pages
     cases

   - First pass to use 'dev_name' instead of ib_device->name

   - Device renaming for RDMA devices"

* tag 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/rdma/rdma: (242 commits)
  IB/mlx5: Add support for extended atomic operations
  RDMA/core: Fix comment for hw stats init for port == 0
  RDMA/core: Refactor ib_register_device() function
  RDMA/core: Fix unwinding flow in case of error to register device
  ib_srp: Remove WARN_ON in srp_terminate_io()
  IB/mlx5: Allow scatter to CQE without global signaled WRs
  IB/mlx5: Verify that driver supports user flags
  IB/mlx5: Support scatter to CQE for DC transport type
  RDMA/drivers: Use core provided API for registering device attributes
  RDMA/core: Allow existing drivers to set one sysfs group per device
  IB/rxe: Remove unnecessary enum values
  RDMA/umad: Use kernel API to allocate umad indexes
  RDMA/uverbs: Use kernel API to allocate uverbs indexes
  RDMA/core: Increase total number of RDMA ports across all devices
  IB/mlx4: Add port and TID to MAD debug print
  IB/mlx4: Enable debug print of SMPs
  RDMA/core: Rename ports_parent to ports_kobj
  RDMA/core: Do not expose unsupported counters
  IB/mlx4: Refer to the device kobject instead of ports_parent
  RDMA/nldev: Allow IB device rename through RDMA netlink
  ...

1 files changed, 36 insertions, 5 deletions

diff --git a/drivers/infiniband/hw/hns/hns_roce_qp.c b/drivers/infiniband/hw/hns/hns_roce_qp.c
index efb7e961ca65..5ebf481a39d9 100644
--- a/drivers/infiniband/hw/hns/hns_roce_qp.c
+++ b/drivers/infiniband/hw/hns/hns_roce_qp.c
@@ -31,6 +31,7 @@
  * SOFTWARE.
  */
 
+#include <linux/pci.h>
 #include <linux/platform_device.h>
 #include <rdma/ib_addr.h>
 #include <rdma/ib_umem.h>
@@ -343,6 +344,7 @@ static int hns_roce_set_user_sq_size(struct hns_roce_dev *hr_dev,
 {
 	u32 roundup_sq_stride = roundup_pow_of_two(hr_dev->caps.max_sq_desc_sz);
 	u8 max_sq_stride = ilog2(roundup_sq_stride);
+	u32 ex_sge_num;
 	u32 page_size;
 	u32 max_cnt;
 
@@ -372,7 +374,18 @@ static int hns_roce_set_user_sq_size(struct hns_roce_dev *hr_dev,
 	if (hr_qp->sq.max_gs > 2)
 		hr_qp->sge.sge_cnt = roundup_pow_of_two(hr_qp->sq.wqe_cnt *
 							(hr_qp->sq.max_gs - 2));
+
+	if ((hr_qp->sq.max_gs > 2) && (hr_dev->pci_dev->revision == 0x20)) {
+		if (hr_qp->sge.sge_cnt > hr_dev->caps.max_extend_sg) {
+			dev_err(hr_dev->dev,
+				"The extended sge cnt error! sge_cnt=%d\n",
+				hr_qp->sge.sge_cnt);
+			return -EINVAL;
+		}
+	}
+
 	hr_qp->sge.sge_shift = 4;
+	ex_sge_num = hr_qp->sge.sge_cnt;
 
 	/* Get buf size, SQ and RQ  are aligned to page_szie */
 	if (hr_dev->caps.max_sq_sg <= 2) {
@@ -386,6 +399,8 @@ static int hns_roce_set_user_sq_size(struct hns_roce_dev *hr_dev,
 					     hr_qp->sq.wqe_shift), PAGE_SIZE);
 	} else {
 		page_size = 1 << (hr_dev->caps.mtt_buf_pg_sz + PAGE_SHIFT);
+		hr_qp->sge.sge_cnt =
+		       max(page_size / (1 << hr_qp->sge.sge_shift), ex_sge_num);
 		hr_qp->buff_size = HNS_ROCE_ALOGN_UP((hr_qp->rq.wqe_cnt <<
 					     hr_qp->rq.wqe_shift), page_size) +
 				   HNS_ROCE_ALOGN_UP((hr_qp->sge.sge_cnt <<
@@ -394,7 +409,7 @@ static int hns_roce_set_user_sq_size(struct hns_roce_dev *hr_dev,
 					     hr_qp->sq.wqe_shift), page_size);
 
 		hr_qp->sq.offset = 0;
-		if (hr_qp->sge.sge_cnt) {
+		if (ex_sge_num) {
 			hr_qp->sge.offset = HNS_ROCE_ALOGN_UP(
 							(hr_qp->sq.wqe_cnt <<
 							hr_qp->sq.wqe_shift),
@@ -465,6 +480,14 @@ static int hns_roce_set_kernel_sq_size(struct hns_roce_dev *hr_dev,
 		hr_qp->sge.sge_shift = 4;
 	}
 
+	if ((hr_qp->sq.max_gs > 2) && hr_dev->pci_dev->revision == 0x20) {
+		if (hr_qp->sge.sge_cnt > hr_dev->caps.max_extend_sg) {
+			dev_err(dev, "The extended sge cnt error! sge_cnt=%d\n",
+				hr_qp->sge.sge_cnt);
+			return -EINVAL;
+		}
+	}
+
 	/* Get buf size, SQ and RQ are aligned to PAGE_SIZE */
 	page_size = 1 << (hr_dev->caps.mtt_buf_pg_sz + PAGE_SHIFT);
 	hr_qp->sq.offset = 0;
@@ -472,6 +495,8 @@ static int hns_roce_set_kernel_sq_size(struct hns_roce_dev *hr_dev,
 				 page_size);
 
 	if (hr_dev->caps.max_sq_sg > 2 && hr_qp->sge.sge_cnt) {
+		hr_qp->sge.sge_cnt = max(page_size/(1 << hr_qp->sge.sge_shift),
+					(u32)hr_qp->sge.sge_cnt);
 		hr_qp->sge.offset = size;
 		size += HNS_ROCE_ALOGN_UP(hr_qp->sge.sge_cnt <<
 					  hr_qp->sge.sge_shift, page_size);
@@ -952,8 +977,8 @@ int hns_roce_modify_qp(struct ib_qp *ibqp, struct ib_qp_attr *attr,
 		}
 	}
 
-	if (!ib_modify_qp_is_ok(cur_state, new_state, ibqp->qp_type, attr_mask,
-				IB_LINK_LAYER_ETHERNET)) {
+	if (!ib_modify_qp_is_ok(cur_state, new_state, ibqp->qp_type,
+				attr_mask)) {
 		dev_err(dev, "ib_modify_qp_is_ok failed\n");
 		goto out;
 	}
@@ -1106,14 +1131,20 @@ int hns_roce_init_qp_table(struct hns_roce_dev *hr_dev)
 {
 	struct hns_roce_qp_table *qp_table = &hr_dev->qp_table;
 	int reserved_from_top = 0;
+	int reserved_from_bot;
 	int ret;
 
 	spin_lock_init(&qp_table->lock);
 	INIT_RADIX_TREE(&hr_dev->qp_table_tree, GFP_ATOMIC);
 
-	/* A port include two SQP, six port total 12 */
+	/* In hw v1, a port include two SQP, six ports total 12 */
+	if (hr_dev->caps.max_sq_sg <= 2)
+		reserved_from_bot = SQP_NUM;
+	else
+		reserved_from_bot = hr_dev->caps.reserved_qps;
+
 	ret = hns_roce_bitmap_init(&qp_table->bitmap, hr_dev->caps.num_qps,
-				   hr_dev->caps.num_qps - 1, SQP_NUM,
+				   hr_dev->caps.num_qps - 1, reserved_from_bot,
 				   reserved_from_top);
 	if (ret) {
 		dev_err(hr_dev->dev, "qp bitmap init failed!error=%d\n",