net/mlx5e: Handle RQ flush in error cases

Add a timeout to avoid an infinite loop waiting for RQ's to flush. This
occurs during AER/EEH and will also happen if the device stops posting
completions due to internal error or reset, or if moving the RQ to the
error state fails. Also cleanup posted receive resources when closing
the RQ.

Fixes: f62b8bb8f2d3 ('net/mlx5: Extend mlx5_core to support ConnectX-4 Ethernet functionality')
Signed-off-by: Daniel Jurgens <danielj@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en.h b/drivers/net/ethernet/mellanox/mlx5/core/en.h
index 244aced4fe70983c65fe45b55b02ffdc3e9b7413..b429591894eb66ebe4d7e6f1503d28611509c9ef 100644 (file)
--- a/drivers/net/ethernet/mellanox/mlx5/core/en.h
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en.h
@@ -191,6 +191,7 @@ struct mlx5e_tstamp {
 enum {
        MLX5E_RQ_STATE_POST_WQES_ENABLE,
        MLX5E_RQ_STATE_UMR_WQE_IN_PROGRESS,
+       MLX5E_RQ_STATE_FLUSH_TIMEOUT,
 };
 
 struct mlx5e_cq {
@@ -220,6 +221,8 @@ typedef void (*mlx5e_fp_handle_rx_cqe)(struct mlx5e_rq *rq,
 typedef int (*mlx5e_fp_alloc_wqe)(struct mlx5e_rq *rq, struct mlx5e_rx_wqe *wqe,
                                  u16 ix);
 
+typedef void (*mlx5e_fp_dealloc_wqe)(struct mlx5e_rq *rq, u16 ix);
+
 struct mlx5e_dma_info {
        struct page     *page;
        dma_addr_t      addr;
@@ -241,6 +244,7 @@ struct mlx5e_rq {
        struct mlx5e_cq        cq;
        mlx5e_fp_handle_rx_cqe handle_rx_cqe;
        mlx5e_fp_alloc_wqe     alloc_wqe;
+       mlx5e_fp_dealloc_wqe   dealloc_wqe;
 
        unsigned long          state;
        int                    ix;
@@ -592,12 +596,15 @@ int mlx5e_napi_poll(struct napi_struct *napi, int budget);
 bool mlx5e_poll_tx_cq(struct mlx5e_cq *cq, int napi_budget);
 int mlx5e_poll_rx_cq(struct mlx5e_cq *cq, int budget);
 void mlx5e_free_tx_descs(struct mlx5e_sq *sq);
+void mlx5e_free_rx_descs(struct mlx5e_rq *rq);
 
 void mlx5e_handle_rx_cqe(struct mlx5e_rq *rq, struct mlx5_cqe64 *cqe);
 void mlx5e_handle_rx_cqe_mpwrq(struct mlx5e_rq *rq, struct mlx5_cqe64 *cqe);
 bool mlx5e_post_rx_wqes(struct mlx5e_rq *rq);
 int mlx5e_alloc_rx_wqe(struct mlx5e_rq *rq, struct mlx5e_rx_wqe *wqe, u16 ix);
 int mlx5e_alloc_rx_mpwqe(struct mlx5e_rq *rq, struct mlx5e_rx_wqe *wqe, u16 ix);
+void mlx5e_dealloc_rx_wqe(struct mlx5e_rq *rq, u16 ix);
+void mlx5e_dealloc_rx_mpwqe(struct mlx5e_rq *rq, u16 ix);
 void mlx5e_post_rx_fragmented_mpwqe(struct mlx5e_rq *rq);
 void mlx5e_complete_rx_linear_mpwqe(struct mlx5e_rq *rq,
                                    struct mlx5_cqe64 *cqe,

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_main.c b/drivers/net/ethernet/mellanox/mlx5/core/en_main.c
index 38c1286abb4e47270db88c9d004d5254a28447ee..103feaba8eec0a1188444cd2ea1279f295eaf2cc 100644 (file)
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_main.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_main.c
@@ -332,6 +332,7 @@ static int mlx5e_create_rq(struct mlx5e_channel *c,
                }
                rq->handle_rx_cqe = mlx5e_handle_rx_cqe_mpwrq;
                rq->alloc_wqe = mlx5e_alloc_rx_mpwqe;
+               rq->dealloc_wqe = mlx5e_dealloc_rx_mpwqe;
 
                rq->mpwqe_stride_sz = BIT(priv->params.mpwqe_log_stride_sz);
                rq->mpwqe_num_strides = BIT(priv->params.mpwqe_log_num_strides);
@@ -347,6 +348,7 @@ static int mlx5e_create_rq(struct mlx5e_channel *c,
                }
                rq->handle_rx_cqe = mlx5e_handle_rx_cqe;
                rq->alloc_wqe = mlx5e_alloc_rx_wqe;
+               rq->dealloc_wqe = mlx5e_dealloc_rx_wqe;
 
                rq->wqe_sz = (priv->params.lro_en) ?
                                priv->params.lro_wqe_sz :
@@ -552,17 +554,25 @@ err_destroy_rq:
 
 static void mlx5e_close_rq(struct mlx5e_rq *rq)
 {
+       int tout = 0;
+       int err;
+
        clear_bit(MLX5E_RQ_STATE_POST_WQES_ENABLE, &rq->state);
        napi_synchronize(&rq->channel->napi); /* prevent mlx5e_post_rx_wqes */
 
-       mlx5e_modify_rq_state(rq, MLX5_RQC_STATE_RDY, MLX5_RQC_STATE_ERR);
-       while (!mlx5_wq_ll_is_empty(&rq->wq))
-               msleep(20);
+       err = mlx5e_modify_rq_state(rq, MLX5_RQC_STATE_RDY, MLX5_RQC_STATE_ERR);
+       while (!mlx5_wq_ll_is_empty(&rq->wq) && !err &&
+              tout++ < MLX5_EN_QP_FLUSH_MAX_ITER)
+               msleep(MLX5_EN_QP_FLUSH_MSLEEP_QUANT);
+
+       if (err || tout == MLX5_EN_QP_FLUSH_MAX_ITER)
+               set_bit(MLX5E_RQ_STATE_FLUSH_TIMEOUT, &rq->state);
 
        /* avoid destroying rq before mlx5e_poll_rx_cq() is done with it */
        napi_synchronize(&rq->channel->napi);
 
        mlx5e_disable_rq(rq);
+       mlx5e_free_rx_descs(rq);
        mlx5e_destroy_rq(rq);
 }
 

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_rx.c b/drivers/net/ethernet/mellanox/mlx5/core/en_rx.c
index 022acc2e8922ae177d26411eee85d900c8150f8d..9f2a16a507e04f8cd9861251ab3d0d5973b23e90 100644 (file)
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_rx.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_rx.c
@@ -212,6 +212,20 @@ err_free_skb:
        return -ENOMEM;
 }
 
+void mlx5e_dealloc_rx_wqe(struct mlx5e_rq *rq, u16 ix)
+{
+       struct sk_buff *skb = rq->skb[ix];
+
+       if (skb) {
+               rq->skb[ix] = NULL;
+               dma_unmap_single(rq->pdev,
+                                *((dma_addr_t *)skb->cb),
+                                rq->wqe_sz,
+                                DMA_FROM_DEVICE);
+               dev_kfree_skb(skb);
+       }
+}
+
 static inline int mlx5e_mpwqe_strides_per_page(struct mlx5e_rq *rq)
 {
        return rq->mpwqe_num_strides >> MLX5_MPWRQ_WQE_PAGE_ORDER;
@@ -574,6 +588,30 @@ int mlx5e_alloc_rx_mpwqe(struct mlx5e_rq *rq, struct mlx5e_rx_wqe *wqe, u16 ix)
        return 0;
 }
 
+void mlx5e_dealloc_rx_mpwqe(struct mlx5e_rq *rq, u16 ix)
+{
+       struct mlx5e_mpw_info *wi = &rq->wqe_info[ix];
+
+       wi->free_wqe(rq, wi);
+}
+
+void mlx5e_free_rx_descs(struct mlx5e_rq *rq)
+{
+       struct mlx5_wq_ll *wq = &rq->wq;
+       struct mlx5e_rx_wqe *wqe;
+       __be16 wqe_ix_be;
+       u16 wqe_ix;
+
+       while (!mlx5_wq_ll_is_empty(wq)) {
+               wqe_ix_be = *wq->tail_next;
+               wqe_ix    = be16_to_cpu(wqe_ix_be);
+               wqe       = mlx5_wq_ll_get_wqe(&rq->wq, wqe_ix);
+               rq->dealloc_wqe(rq, wqe_ix);
+               mlx5_wq_ll_pop(&rq->wq, wqe_ix_be,
+                              &wqe->next.next_wqe_index);
+       }
+}
+
 #define RQ_CANNOT_POST(rq) \
                (!test_bit(MLX5E_RQ_STATE_POST_WQES_ENABLE, &rq->state) || \
                 test_bit(MLX5E_RQ_STATE_UMR_WQE_IN_PROGRESS, &rq->state))
@@ -878,6 +916,9 @@ int mlx5e_poll_rx_cq(struct mlx5e_cq *cq, int budget)
        struct mlx5e_rq *rq = container_of(cq, struct mlx5e_rq, cq);
        int work_done = 0;
 
+       if (unlikely(test_bit(MLX5E_RQ_STATE_FLUSH_TIMEOUT, &rq->state)))
+               return 0;
+
        if (cq->decmprs_left)
                work_done += mlx5e_decompress_cqes_cont(rq, cq, 0, budget);