IB/mlx5: Implement fragmented completion queue (CQ)

The current implementation of create CQ requires contiguous
memory, such requirement is problematic once the memory is
fragmented or the system is low in memory, it causes for
failures in dma_zalloc_coherent().

This patch implements new scheme of fragmented CQ to overcome
this issue by introducing new type: 'struct mlx5_frag_buf_ctrl'
to allocate fragmented buffers, rather than contiguous ones.

Base the Completion Queues (CQs) on this new fragmented buffer.

It fixes following crashes:
kworker/29:0: page allocation failure: order:6, mode:0x80d0
CPU: 29 PID: 8374 Comm: kworker/29:0 Tainted: G OE 3.10.0
Workqueue: ib_cm cm_work_handler [ib_cm]
Call Trace:
[<>] dump_stack+0x19/0x1b
[<>] warn_alloc_failed+0x110/0x180
[<>] __alloc_pages_slowpath+0x6b7/0x725
[<>] __alloc_pages_nodemask+0x405/0x420
[<>] dma_generic_alloc_coherent+0x8f/0x140
[<>] x86_swiotlb_alloc_coherent+0x21/0x50
[<>] mlx5_dma_zalloc_coherent_node+0xad/0x110 [mlx5_core]
[<>] ? mlx5_db_alloc_node+0x69/0x1b0 [mlx5_core]
[<>] mlx5_buf_alloc_node+0x3e/0xa0 [mlx5_core]
[<>] mlx5_buf_alloc+0x14/0x20 [mlx5_core]
[<>] create_cq_kernel+0x90/0x1f0 [mlx5_ib]
[<>] mlx5_ib_create_cq+0x3b0/0x4e0 [mlx5_ib]

Signed-off-by: Yonatan Cohen <yonatanc@mellanox.com>
Reviewed-by: Tariq Toukan <tariqt@mellanox.com>
Signed-off-by: Leon Romanovsky <leon@kernel.org>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>

4 files changed, 45 insertions, 43 deletions

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/alloc.c b/drivers/net/ethernet/mellanox/mlx5/core/alloc.c
index 47239bf7bf43..323ffe8bf7e4 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/alloc.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/alloc.c
@@ -71,19 +71,24 @@ static void *mlx5_dma_zalloc_coherent_node(struct mlx5_core_dev *dev,
 }
 
 int mlx5_buf_alloc_node(struct mlx5_core_dev *dev, int size,
-			struct mlx5_buf *buf, int node)
+			struct mlx5_frag_buf *buf, int node)
 {
 	dma_addr_t t;
 
 	buf->size = size;
 	buf->npages       = 1;
 	buf->page_shift   = (u8)get_order(size) + PAGE_SHIFT;
-	buf->direct.buf   = mlx5_dma_zalloc_coherent_node(dev, size,
-							  &t, node);
-	if (!buf->direct.buf)
+
+	buf->frags = kzalloc(sizeof(*buf->frags), GFP_KERNEL);
+	if (!buf->frags)
 		return -ENOMEM;
 
-	buf->direct.map = t;
+	buf->frags->buf   = mlx5_dma_zalloc_coherent_node(dev, size,
+							  &t, node);
+	if (!buf->frags->buf)
+		goto err_out;
+
+	buf->frags->map = t;
 
 	while (t & ((1 << buf->page_shift) - 1)) {
 		--buf->page_shift;
@@ -91,18 +96,24 @@ int mlx5_buf_alloc_node(struct mlx5_core_dev *dev, int size,
 	}
 
 	return 0;
+err_out:
+	kfree(buf->frags);
+	return -ENOMEM;
 }
 
-int mlx5_buf_alloc(struct mlx5_core_dev *dev, int size, struct mlx5_buf *buf)
+int mlx5_buf_alloc(struct mlx5_core_dev *dev,
+		   int size, struct mlx5_frag_buf *buf)
 {
 	return mlx5_buf_alloc_node(dev, size, buf, dev->priv.numa_node);
 }
-EXPORT_SYMBOL_GPL(mlx5_buf_alloc);
+EXPORT_SYMBOL(mlx5_buf_alloc);
 
-void mlx5_buf_free(struct mlx5_core_dev *dev, struct mlx5_buf *buf)
+void mlx5_buf_free(struct mlx5_core_dev *dev, struct mlx5_frag_buf *buf)
 {
-	dma_free_coherent(&dev->pdev->dev, buf->size, buf->direct.buf,
-			  buf->direct.map);
+	dma_free_coherent(&dev->pdev->dev, buf->size, buf->frags->buf,
+			  buf->frags->map);
+
+	kfree(buf->frags);
 }
 EXPORT_SYMBOL_GPL(mlx5_buf_free);
 
@@ -147,6 +158,7 @@ err_free_buf:
 err_out:
 	return -ENOMEM;
 }
+EXPORT_SYMBOL_GPL(mlx5_frag_buf_alloc_node);
 
 void mlx5_frag_buf_free(struct mlx5_core_dev *dev, struct mlx5_frag_buf *buf)
 {
@@ -162,6 +174,7 @@ void mlx5_frag_buf_free(struct mlx5_core_dev *dev, struct mlx5_frag_buf *buf)
 	}
 	kfree(buf->frags);
 }
+EXPORT_SYMBOL_GPL(mlx5_frag_buf_free);
 
 static struct mlx5_db_pgdir *mlx5_alloc_db_pgdir(struct mlx5_core_dev *dev,
 						 int node)
@@ -275,13 +288,13 @@ void mlx5_db_free(struct mlx5_core_dev *dev, struct mlx5_db *db)
 }
 EXPORT_SYMBOL_GPL(mlx5_db_free);
 
-void mlx5_fill_page_array(struct mlx5_buf *buf, __be64 *pas)
+void mlx5_fill_page_array(struct mlx5_frag_buf *buf, __be64 *pas)
 {
 	u64 addr;
 	int i;
 
 	for (i = 0; i < buf->npages; i++) {
-		addr = buf->direct.map + (i << buf->page_shift);
+		addr = buf->frags->map + (i << buf->page_shift);
 
 		pas[i] = cpu_to_be64(addr);
 	}
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_rx.c b/drivers/net/ethernet/mellanox/mlx5/core/en_rx.c
index 0d4bb0688faa..80b84f6af2a1 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_rx.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_rx.c
@@ -52,7 +52,7 @@ static inline bool mlx5e_rx_hw_stamp(struct hwtstamp_config *config)
 static inline void mlx5e_read_cqe_slot(struct mlx5e_cq *cq, u32 cqcc,
 				       void *data)
 {
-	u32 ci = cqcc & cq->wq.sz_m1;
+	u32 ci = cqcc & cq->wq.fbc.sz_m1;
 
 	memcpy(data, mlx5_cqwq_get_wqe(&cq->wq, ci), sizeof(struct mlx5_cqe64));
 }
@@ -74,9 +74,10 @@ static inline void mlx5e_read_mini_arr_slot(struct mlx5e_cq *cq, u32 cqcc)
 
 static inline void mlx5e_cqes_update_owner(struct mlx5e_cq *cq, u32 cqcc, int n)
 {
-	u8 op_own = (cqcc >> cq->wq.log_sz) & 1;
-	u32 wq_sz = 1 << cq->wq.log_sz;
-	u32 ci = cqcc & cq->wq.sz_m1;
+	struct mlx5_frag_buf_ctrl *fbc = &cq->wq.fbc;
+	u8 op_own = (cqcc >> fbc->log_sz) & 1;
+	u32 wq_sz = 1 << fbc->log_sz;
+	u32 ci = cqcc & fbc->sz_m1;
 	u32 ci_top = min_t(u32, wq_sz, ci + n);
 
 	for (; ci < ci_top; ci++, n--) {
@@ -101,7 +102,7 @@ static inline void mlx5e_decompress_cqe(struct mlx5e_rq *rq,
 	cq->title.byte_cnt     = cq->mini_arr[cq->mini_arr_idx].byte_cnt;
 	cq->title.check_sum    = cq->mini_arr[cq->mini_arr_idx].checksum;
 	cq->title.op_own      &= 0xf0;
-	cq->title.op_own      |= 0x01 & (cqcc >> cq->wq.log_sz);
+	cq->title.op_own      |= 0x01 & (cqcc >> cq->wq.fbc.log_sz);
 	cq->title.wqe_counter  = cpu_to_be16(cq->decmprs_wqe_counter);
 
 	if (rq->wq_type == MLX5_WQ_TYPE_LINKED_LIST_STRIDING_RQ)
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/wq.c b/drivers/net/ethernet/mellanox/mlx5/core/wq.c
index 6bcfc25350f5..ea66448ba365 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/wq.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/wq.c
@@ -41,7 +41,7 @@ u32 mlx5_wq_cyc_get_size(struct mlx5_wq_cyc *wq)
 
 u32 mlx5_cqwq_get_size(struct mlx5_cqwq *wq)
 {
-	return wq->sz_m1 + 1;
+	return wq->fbc.sz_m1 + 1;
 }
 
 u32 mlx5_wq_ll_get_size(struct mlx5_wq_ll *wq)
@@ -62,7 +62,7 @@ static u32 mlx5_wq_qp_get_byte_size(struct mlx5_wq_qp *wq)
 
 static u32 mlx5_cqwq_get_byte_size(struct mlx5_cqwq *wq)
 {
-	return mlx5_cqwq_get_size(wq) << wq->log_stride;
+	return mlx5_cqwq_get_size(wq) << wq->fbc.log_stride;
 }
 
 static u32 mlx5_wq_ll_get_byte_size(struct mlx5_wq_ll *wq)
@@ -92,7 +92,7 @@ int mlx5_wq_cyc_create(struct mlx5_core_dev *mdev, struct mlx5_wq_param *param,
 		goto err_db_free;
 	}
 
-	wq->buf = wq_ctrl->buf.direct.buf;
+	wq->buf = wq_ctrl->buf.frags->buf;
 	wq->db  = wq_ctrl->db.db;
 
 	wq_ctrl->mdev = mdev;
@@ -130,7 +130,7 @@ int mlx5_wq_qp_create(struct mlx5_core_dev *mdev, struct mlx5_wq_param *param,
 		goto err_db_free;
 	}
 
-	wq->rq.buf = wq_ctrl->buf.direct.buf;
+	wq->rq.buf = wq_ctrl->buf.frags->buf;
 	wq->sq.buf = wq->rq.buf + mlx5_wq_cyc_get_byte_size(&wq->rq);
 	wq->rq.db  = &wq_ctrl->db.db[MLX5_RCV_DBR];
 	wq->sq.db  = &wq_ctrl->db.db[MLX5_SND_DBR];
@@ -151,11 +151,7 @@ int mlx5_cqwq_create(struct mlx5_core_dev *mdev, struct mlx5_wq_param *param,
 {
 	int err;
 
-	wq->log_stride	= 6 + MLX5_GET(cqc, cqc, cqe_sz);
-	wq->log_sz	= MLX5_GET(cqc, cqc, log_cq_size);
-	wq->sz_m1	= (1 << wq->log_sz) - 1;
-	wq->log_frag_strides = PAGE_SHIFT - wq->log_stride;
-	wq->frag_sz_m1	= (1 << wq->log_frag_strides) - 1;
+	mlx5_core_init_cq_frag_buf(&wq->fbc, cqc);
 
 	err = mlx5_db_alloc_node(mdev, &wq_ctrl->db, param->db_numa_node);
 	if (err) {
@@ -172,7 +168,7 @@ int mlx5_cqwq_create(struct mlx5_core_dev *mdev, struct mlx5_wq_param *param,
 		goto err_db_free;
 	}
 
-	wq->frag_buf = wq_ctrl->frag_buf;
+	wq->fbc.frag_buf = wq_ctrl->frag_buf;
 	wq->db  = wq_ctrl->db.db;
 
 	wq_ctrl->mdev = mdev;
@@ -209,7 +205,7 @@ int mlx5_wq_ll_create(struct mlx5_core_dev *mdev, struct mlx5_wq_param *param,
 		goto err_db_free;
 	}
 
-	wq->buf = wq_ctrl->buf.direct.buf;
+	wq->buf = wq_ctrl->buf.frags->buf;
 	wq->db  = wq_ctrl->db.db;
 
 	for (i = 0; i < wq->sz_m1; i++) {
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/wq.h b/drivers/net/ethernet/mellanox/mlx5/core/wq.h
index 718589d0cec2..fca90b94596d 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/wq.h
+++ b/drivers/net/ethernet/mellanox/mlx5/core/wq.h
@@ -45,7 +45,7 @@ struct mlx5_wq_param {
 
 struct mlx5_wq_ctrl {
 	struct mlx5_core_dev	*mdev;
-	struct mlx5_buf		buf;
+	struct mlx5_frag_buf	buf;
 	struct mlx5_db		db;
 };
 
@@ -68,14 +68,9 @@ struct mlx5_wq_qp {
 };
 
 struct mlx5_cqwq {
-	struct mlx5_frag_buf	frag_buf;
-	__be32			*db;
-	u32			sz_m1;
-	u32			frag_sz_m1;
-	u32			cc; /* consumer counter */
-	u8			log_sz;
-	u8			log_stride;
-	u8			log_frag_strides;
+	struct mlx5_frag_buf_ctrl fbc;
+	__be32			  *db;
+	u32			  cc; /* consumer counter */
 };
 
 struct mlx5_wq_ll {
@@ -131,20 +126,17 @@ static inline int mlx5_wq_cyc_cc_bigger(u16 cc1, u16 cc2)
 
 static inline u32 mlx5_cqwq_get_ci(struct mlx5_cqwq *wq)
 {
-	return wq->cc & wq->sz_m1;
+	return wq->cc & wq->fbc.sz_m1;
 }
 
 static inline void *mlx5_cqwq_get_wqe(struct mlx5_cqwq *wq, u32 ix)
 {
-	unsigned int frag = (ix >> wq->log_frag_strides);
-
-	return wq->frag_buf.frags[frag].buf +
-		((wq->frag_sz_m1 & ix) << wq->log_stride);
+	return mlx5_frag_buf_get_wqe(&wq->fbc, ix);
 }
 
 static inline u32 mlx5_cqwq_get_wrap_cnt(struct mlx5_cqwq *wq)
 {
-	return wq->cc >> wq->log_sz;
+	return wq->cc >> wq->fbc.log_sz;
 }
 
 static inline void mlx5_cqwq_pop(struct mlx5_cqwq *wq)