[PATCH rdma-next] RDMA/mlx5: print wc status on CQE error and dump needed

[PATCH rdma-next] RDMA/mlx5: print wc status on CQE error and dump needed

[Dust Li]

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:
[166755.330649] infiniband mlx5_0: mlx5_handle_error_cqe:333:(pid 0): WC error: 10, message: remote access error
[166755.332323] infiniband mlx5_0: dump_cqe:272:(pid 0): dump error cqe
[166755.332944] 00000000: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
[166755.333574] 00000010: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
[166755.334202] 00000020: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
[166755.334837] 00000030: 00 00 00 00 00 00 88 13 08 03 61 b3 1e a1 42 d3

Signed-off-by: Dust Li <dust.li@linux.alibaba.com>
---
 drivers/infiniband/hw/mlx5/cq.c | 5 ++++-
 1 file changed, 4 insertions(+), 1 deletion(-)

diff --git a/drivers/infiniband/hw/mlx5/cq.c b/drivers/infiniband/hw/mlx5/cq.c
index a190fb581591..66dfadb96c66 100644
--- a/drivers/infiniband/hw/mlx5/cq.c
+++ b/drivers/infiniband/hw/mlx5/cq.c
@@ -328,8 +328,11 @@ 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,
-- 
2.19.1.3.ge56e4f7

Re: [PATCH rdma-next] RDMA/mlx5: print wc status on CQE error and dump needed

[Leon Romanovsky]

On Mon, Dec 27, 2021 at 08:38:06PM +0800, Dust Li wrote:
> 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:
> [166755.330649] infiniband mlx5_0: mlx5_handle_error_cqe:333:(pid 0): WC error: 10, message: remote access error
> [166755.332323] infiniband mlx5_0: dump_cqe:272:(pid 0): dump error cqe
> [166755.332944] 00000000: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
> [166755.333574] 00000010: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
> [166755.334202] 00000020: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
> [166755.334837] 00000030: 00 00 00 00 00 00 88 13 08 03 61 b3 1e a1 42 d3
> 
> Signed-off-by: Dust Li <dust.li@linux.alibaba.com>
> ---
>  drivers/infiniband/hw/mlx5/cq.c | 5 ++++-
>  1 file changed, 4 insertions(+), 1 deletion(-)
> 

Thanks,
Acked-by: Leon Romanovsky <leonro@nvidia.com>

Re: [PATCH rdma-next] RDMA/mlx5: print wc status on CQE error and dump needed

[Jason Gunthorpe]

On Mon, Dec 27, 2021 at 08:38:06PM +0800, Dust Li wrote:
> 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:
> [166755.330649] infiniband mlx5_0: mlx5_handle_error_cqe:333:(pid 0): WC error: 10, message: remote access error
> [166755.332323] infiniband mlx5_0: dump_cqe:272:(pid 0): dump error cqe
> [166755.332944] 00000000: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
> [166755.333574] 00000010: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
> [166755.334202] 00000020: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
> [166755.334837] 00000030: 00 00 00 00 00 00 88 13 08 03 61 b3 1e a1 42 d3
> 
> Signed-off-by: Dust Li <dust.li@linux.alibaba.com>
> Acked-by: Leon Romanovsky <leonro@nvidia.com>
> ---
>  drivers/infiniband/hw/mlx5/cq.c | 5 ++++-
>  1 file changed, 4 insertions(+), 1 deletion(-)

Applied to for-next, thanks

Jason