Commit a7ad9dde authored by Dust Li's avatar Dust Li Committed by Jason Gunthorpe
Browse files

RDMA/mlx5: Print wc status on CQE error and dump needed 

mlx5_handle_error_cqe() only dump the content of the CQE which is raw hex
data, and not straighforward for debug.  Print WC status message when we
got CQE error and dump is need.

Here is an example of how the dmesg log looks like with this:

 infiniband mlx5_0: mlx5_handle_error_cqe:333:(pid 0): WC error: 10, message: remote access error
 infiniband mlx5_0: dump_cqe:272:(pid 0): dump error cqe
 00000000: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
 00000010: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
 00000020: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
 00000030: 00 00 00 00 00 00 88 13 08 03 61 b3 1e a1 42 d3

Link: https://lore.kernel.org/r/20211227123806.47530-1-dust.li@linux.alibaba.com


Signed-off-by: default avatarDust Li <dust.li@linux.alibaba.com>
Acked-by: default avatarLeon Romanovsky <leonro@nvidia.com>
Signed-off-by: default avatarJason Gunthorpe <jgg@nvidia.com>
parent 8d1cfb88

drivers/infiniband/hw/mlx5/cq.c

+4 −1
Original line number Diff line number Diff line
@@ -328,9 +328,12 @@ static void mlx5_handle_error_cqe(struct mlx5_ib_dev *dev, 
	}



	wc->vendor_err = cqe->vendor_err_synd;

	if (dump)

	if (dump) {

		mlx5_ib_warn(dev, "WC error: %d, Message: %s\n", wc->status,

			     ib_wc_status_msg(wc->status));

		dump_cqe(dev, cqe);

	}

}



static void handle_atomics(struct mlx5_ib_qp *qp, struct mlx5_cqe64 *cqe64,

			   u16 tail, u16 head)