Merge tag 'v6.6' into rdma.git for-next

Resolve conflict by taking the spin_lock hunk from for-next:

 https://lore.kernel.org/r/20230928113851.5197a1ec@canb.auug.org.au

Required for the next patch.

Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>

8 files changed, 33 insertions, 12 deletions

diff --git a/drivers/infiniband/hw/bnxt_re/ib_verbs.c b/drivers/infiniband/hw/bnxt_re/ib_verbs.c
index 0848c2c2ffcf..faa88d12ee86 100644
--- a/drivers/infiniband/hw/bnxt_re/ib_verbs.c
+++ b/drivers/infiniband/hw/bnxt_re/ib_verbs.c
@@ -910,6 +910,10 @@ int bnxt_re_destroy_qp(struct ib_qp *ib_qp, struct ib_udata *udata)
 	list_del(&qp->list);
 	mutex_unlock(&rdev->qp_lock);
 	atomic_dec(&rdev->stats.res.qp_count);
+	if (qp->qplib_qp.type == CMDQ_CREATE_QP_TYPE_RC)
+		atomic_dec(&rdev->stats.res.rc_qp_count);
+	else if (qp->qplib_qp.type == CMDQ_CREATE_QP_TYPE_UD)
+		atomic_dec(&rdev->stats.res.ud_qp_count);
 
 	ib_umem_release(qp->rumem);
 	ib_umem_release(qp->sumem);
diff --git a/drivers/infiniband/hw/bnxt_re/qplib_rcfw.c b/drivers/infiniband/hw/bnxt_re/qplib_rcfw.c
index c8c4017fe405..e47b4ca64d33 100644
--- a/drivers/infiniband/hw/bnxt_re/qplib_rcfw.c
+++ b/drivers/infiniband/hw/bnxt_re/qplib_rcfw.c
@@ -665,7 +665,6 @@ static int bnxt_qplib_process_qp_event(struct bnxt_qplib_rcfw *rcfw,
 		blocked = cookie & RCFW_CMD_IS_BLOCKING;
 		cookie &= RCFW_MAX_COOKIE_VALUE;
 		crsqe = &rcfw->crsqe_tbl[cookie];
-		crsqe->is_in_used = false;
 
 		if (WARN_ONCE(test_bit(FIRMWARE_STALL_DETECTED,
 				       &rcfw->cmdq.flags),
@@ -681,8 +680,14 @@ static int bnxt_qplib_process_qp_event(struct bnxt_qplib_rcfw *rcfw,
 			atomic_dec(&rcfw->timeout_send);
 
 		if (crsqe->is_waiter_alive) {
-			if (crsqe->resp)
+			if (crsqe->resp) {
 				memcpy(crsqe->resp, qp_event, sizeof(*qp_event));
+				/* Insert write memory barrier to ensure that
+				 * response data is copied before clearing the
+				 * flags
+				 */
+				smp_wmb();
+			}
 			if (!blocked)
 				wait_cmds++;
 		}
@@ -694,6 +699,8 @@ static int bnxt_qplib_process_qp_event(struct bnxt_qplib_rcfw *rcfw,
 		if (!is_waiter_alive)
 			crsqe->resp = NULL;
 
+		crsqe->is_in_used = false;
+
 		hwq->cons += req_size;
 
 		/* This is a case to handle below scenario -
diff --git a/drivers/infiniband/hw/cxgb4/cm.c b/drivers/infiniband/hw/cxgb4/cm.c
index ced615b5ea09..040ba2224f9f 100644
--- a/drivers/infiniband/hw/cxgb4/cm.c
+++ b/drivers/infiniband/hw/cxgb4/cm.c
@@ -1965,6 +1965,9 @@ static int send_fw_act_open_req(struct c4iw_ep *ep, unsigned int atid)
 	int win;
 
 	skb = get_skb(NULL, sizeof(*req), GFP_KERNEL);
+	if (!skb)
+		return -ENOMEM;
+
 	req = __skb_put_zero(skb, sizeof(*req));
 	req->op_compl = htonl(WR_OP_V(FW_OFLD_CONNECTION_WR));
 	req->len16_pkd = htonl(FW_WR_LEN16_V(DIV_ROUND_UP(sizeof(*req), 16)));
diff --git a/drivers/infiniband/hw/erdma/erdma_verbs.c b/drivers/infiniband/hw/erdma/erdma_verbs.c
index dcccb6015232..c317947563fb 100644
--- a/drivers/infiniband/hw/erdma/erdma_verbs.c
+++ b/drivers/infiniband/hw/erdma/erdma_verbs.c
@@ -133,8 +133,8 @@ static int create_qp_cmd(struct erdma_ucontext *uctx, struct erdma_qp *qp)
 static int regmr_cmd(struct erdma_dev *dev, struct erdma_mr *mr)
 {
 	struct erdma_pd *pd = to_epd(mr->ibmr.pd);
+	u32 mtt_level = ERDMA_MR_MTT_0LEVEL;
 	struct erdma_cmdq_reg_mr_req req;
-	u32 mtt_level;
 
 	erdma_cmdq_build_reqhdr(&req.hdr, CMDQ_SUBMOD_RDMA, CMDQ_OPCODE_REG_MR);
 
@@ -147,10 +147,9 @@ static int regmr_cmd(struct erdma_dev *dev, struct erdma_mr *mr)
 			req.phy_addr[0] = sg_dma_address(mr->mem.mtt->sglist);
 			mtt_level = mr->mem.mtt->level;
 		}
-	} else {
+	} else if (mr->type != ERDMA_MR_TYPE_DMA) {
 		memcpy(req.phy_addr, mr->mem.mtt->buf,
 		       MTT_SIZE(mr->mem.page_cnt));
-		mtt_level = ERDMA_MR_MTT_0LEVEL;
 	}
 
 	req.cfg0 = FIELD_PREP(ERDMA_CMD_MR_VALID_MASK, mr->valid) |
@@ -655,7 +654,7 @@ static struct erdma_mtt *erdma_create_scatter_mtt(struct erdma_dev *dev,
 
 	mtt = kzalloc(sizeof(*mtt), GFP_KERNEL);
 	if (!mtt)
-		return NULL;
+		return ERR_PTR(-ENOMEM);
 
 	mtt->size = ALIGN(size, PAGE_SIZE);
 	mtt->buf = vzalloc(mtt->size);
diff --git a/drivers/infiniband/hw/mlx4/sysfs.c b/drivers/infiniband/hw/mlx4/sysfs.c
index 24ee79aa2122..88f534cf690e 100644
--- a/drivers/infiniband/hw/mlx4/sysfs.c
+++ b/drivers/infiniband/hw/mlx4/sysfs.c
@@ -223,7 +223,7 @@ void del_sysfs_port_mcg_attr(struct mlx4_ib_dev *device, int port_num,
 static int add_port_entries(struct mlx4_ib_dev *device, int port_num)
 {
 	int i;
-	char buff[11];
+	char buff[12];
 	struct mlx4_ib_iov_port *port = NULL;
 	int ret = 0 ;
 	struct ib_port_attr attr;
diff --git a/drivers/infiniband/hw/mlx5/fs.c b/drivers/infiniband/hw/mlx5/fs.c
index 1e419e080b53..520034acf73a 100644
--- a/drivers/infiniband/hw/mlx5/fs.c
+++ b/drivers/infiniband/hw/mlx5/fs.c
@@ -2470,8 +2470,8 @@ destroy_res:
 	mlx5_steering_anchor_destroy_res(ft_prio);
 put_flow_table:
 	put_flow_table(dev, ft_prio, true);
-	mutex_unlock(&dev->flow_db->lock);
 free_obj:
+	mutex_unlock(&dev->flow_db->lock);
 	kfree(obj);
 
 	return err;
diff --git a/drivers/infiniband/hw/mlx5/main.c b/drivers/infiniband/hw/mlx5/main.c
index ec7c45272764..e39e4b06712a 100644
--- a/drivers/infiniband/hw/mlx5/main.c
+++ b/drivers/infiniband/hw/mlx5/main.c
@@ -2089,7 +2089,7 @@ static inline char *mmap_cmd2str(enum mlx5_ib_mmap_cmd cmd)
 	case MLX5_IB_MMAP_DEVICE_MEM:
 		return "Device Memory";
 	default:
-		return NULL;
+		return "Unknown";
 	}
 }
 
diff --git a/drivers/infiniband/hw/mlx5/mr.c b/drivers/infiniband/hw/mlx5/mr.c
index b0fa2d644973..1d6c54a53df6 100644
--- a/drivers/infiniband/hw/mlx5/mr.c
+++ b/drivers/infiniband/hw/mlx5/mr.c
@@ -239,7 +239,8 @@ static int get_mkc_octo_size(unsigned int access_mode, unsigned int ndescs)
 
 static void set_cache_mkc(struct mlx5_cache_ent *ent, void *mkc)
 {
-	set_mkc_access_pd_addr_fields(mkc, 0, 0, ent->dev->umrc.pd);
+	set_mkc_access_pd_addr_fields(mkc, ent->rb_key.access_flags, 0,
+				      ent->dev->umrc.pd);
 	MLX5_SET(mkc, mkc, free, 1);
 	MLX5_SET(mkc, mkc, umr_en, 1);
 	MLX5_SET(mkc, mkc, access_mode_1_0, ent->rb_key.access_mode & 0x3);
@@ -998,19 +999,26 @@ void mlx5_mkey_cache_cleanup(struct mlx5_ib_dev *dev)
 	if (!dev->cache.wq)
 		return;
 
-	cancel_delayed_work_sync(&dev->cache.remove_ent_dwork);
 	mutex_lock(&dev->cache.rb_lock);
 	for (node = rb_first(root); node; node = rb_next(node)) {
 		ent = rb_entry(node, struct mlx5_cache_ent, node);
 		spin_lock_irq(&ent->mkeys_queue.lock);
 		ent->disabled = true;
 		spin_unlock_irq(&ent->mkeys_queue.lock);
-		cancel_delayed_work_sync(&ent->dwork);
 	}
+	mutex_unlock(&dev->cache.rb_lock);
+
+	/*
+	 * After all entries are disabled and will not reschedule on WQ,
+	 * flush it and all async commands.
+	 */
+	flush_workqueue(dev->cache.wq);
 
 	mlx5_mkey_cache_debugfs_cleanup(dev);
 	mlx5_cmd_cleanup_async_ctx(&dev->async_ctx);
 
+	/* At this point all entries are disabled and have no concurrent work. */
+	mutex_lock(&dev->cache.rb_lock);
 	node = rb_first(root);
 	while (node) {
 		ent = rb_entry(node, struct mlx5_cache_ent, node);