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>

1 files changed, 7 insertions, 11 deletions

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++) {