Hi,
> -----Original Message-----
> From: Dekel Peled <dek...@mellanox.com>
> Sent: Monday, July 27, 2020 11:51 AM
> To: Matan Azrad <ma...@mellanox.com>; Slava Ovsiienko
> <viachesl...@mellanox.com>; Raslan Darawsheh <rasl...@mellanox.com>
> Cc: dev@dpdk.org
> Subject: [PATCH] net/mlx5: fix CQ interrupt handling and cleanup
>
> Recent patch added creation of Rx CQ using DevX API.
> The reading of events from DevX channel was not done correctly.
> This patch fixes the event reading, using the correct data structure.
> Cleanup after CQ creation, in case of error, is also updated.
>
> Fixes: 08d1838f645a ("net/mlx5: implement CQ for Rx using DevX API")
>
> Signed-off-by: Dekel Peled <dek...@mellanox.com>
> Acked-by: Viacheslav Ovsiienko <viachesl...@mellanox.com>
> ---
> drivers/net/mlx5/mlx5_rxq.c | 14 ++++++++++----
> 1 file changed, 10 insertions(+), 4 deletions(-)
>
> diff --git a/drivers/net/mlx5/mlx5_rxq.c b/drivers/net/mlx5/mlx5_rxq.c
> index c78e522..79eb8f8 100644
> --- a/drivers/net/mlx5/mlx5_rxq.c
> +++ b/drivers/net/mlx5/mlx5_rxq.c
> @@ -1193,12 +1193,16 @@
> mlx5_glue->ack_cq_events(rxq_obj->ibv_cq, 1);
> } else if (rxq_obj->type == MLX5_RXQ_OBJ_TYPE_DEVX_RQ) {
> #ifdef HAVE_IBV_DEVX_EVENT
> - struct mlx5dv_devx_async_event_hdr *event_data = NULL;
> + union {
> + struct mlx5dv_devx_async_event_hdr event_resp;
> + uint8_t buf[sizeof(struct
> mlx5dv_devx_async_event_hdr)
> + + 128];
> + } out;
>
> ret = mlx5_glue->devx_get_event
> - (rxq_obj->devx_channel, event_data,
> - sizeof(struct
> mlx5dv_devx_async_event_hdr));
> - if (ret < 0 || event_data->cookie !=
> + (rxq_obj->devx_channel, &out.event_resp,
> + sizeof(out.buf));
> + if (ret < 0 || out.event_resp.cookie !=
> (uint64_t)(uintptr_t)rxq_obj->devx_cq)
> goto exit;
> #endif /* HAVE_IBV_DEVX_EVENT */
> @@ -1646,6 +1650,8 @@
> memset((void *)(uintptr_t)rxq_data->cqes, 0xFF, cq_size);
> return cq_obj;
> error:
> + if (cq_obj)
> + mlx5_devx_cmd_destroy(cq_obj);
> rxq_release_devx_cq_resources(rxq_ctrl);
> return NULL;
> }
> --
> 1.8.3.1
Patch applied to next-net-mlx,
Kindest regards,
Raslan Darawsheh
