|SUCCESS| [v2] common/mlx5: add ConnectX-10 SmartNIC

dpdklab at iol.unh.edu dpdklab at iol.unh.edu
Mon Apr 20 15:57:37 CEST 2026


Test-Label: ai-code-review
Test-Status: SUCCESS
https://dpdk.org/patch/163218

_AI Code Review_

=== Patch Review: series.patch (via Claude) ===
Review date: 2026-04-20

git sha: (Not provided - patch from patchwork)
Author: Raslan Darawsheh <rasland at nvidia.com>

Subject: [PATCH v2] common/mlx5: add ConnectX-10 SmartNIC

This patch adds the ConnectX-10 SmartNIC device ID to the list of NVIDIA
devices that run the mlx5 drivers. The device is still in development
stage.

> diff --git a/drivers/common/mlx5/mlx5_common.h b/drivers/common/mlx5/mlx5_common.h
> index 47fcc3fb06..3e66c9e6c8 100644
> --- a/drivers/common/mlx5/mlx5_common.h
> +++ b/drivers/common/mlx5/mlx5_common.h
> @@ -156,6 +156,7 @@ enum {
>  	PCI_DEVICE_ID_MELLANOX_CONNECTX8 = 0x1023,
>  	PCI_DEVICE_ID_MELLANOX_CONNECTX9 = 0x1025,
>  	PCI_DEVICE_ID_MELLANOX_BLUEFIELD4 = 0xa2df,
> +	PCI_DEVICE_ID_MELLANOX_CONNECTX10 = 0x1027,
>  };

Does this enum maintain a deliberate ordering where device IDs should be
sorted by value? The new CONNECTX10 entry (0x1027) appears after
BLUEFIELD4 (0xa2df) but has a numerically lower value. Was the
placement intentional, or should it be placed between CONNECTX9
(0x1025) and BLUEFIELD4 to maintain numerical ordering?

[ ... ]

> diff --git a/drivers/crypto/mlx5/mlx5_crypto.c b/drivers/crypto/mlx5/mlx5_crypto.c
> index 240e8bb858..dd0aabb6d7 100644
> --- a/drivers/crypto/mlx5/mlx5_crypto.c
> +++ b/drivers/crypto/mlx5/mlx5_crypto.c
> @@ -495,6 +495,10 @@ static const struct rte_pci_id mlx5_crypto_pci_id_map[] = {
>  			RTE_PCI_DEVICE(PCI_VENDOR_ID_MELLANOX,
>  					PCI_DEVICE_ID_MELLANOX_BLUEFIELD4)
>  		},
> +		{
> +			RTE_PCI_DEVICE(PCI_VENDOR_ID_MELLANOX,
> +					PCI_DEVICE_ID_MELLANOX_CONNECTX10)
> +		},
>  		{
>  			.vendor_id = 0
>  		}

Is there whitespace aligned at the start of the line in this hunk? The
RTE_PCI_DEVICE and PCI_DEVICE_ID_MELLANOX_CONNECTX10 lines appear to
use tabs followed by spaces. Can you verify that the indentation uses
only tabs where appropriate, per the DPDK coding style?

> diff --git a/drivers/net/mlx5/mlx5.c b/drivers/net/mlx5/mlx5.c
> index c24c9b306d..f190654756 100644
> --- a/drivers/net/mlx5/mlx5.c
> +++ b/drivers/net/mlx5/mlx5.c
> @@ -3925,6 +3925,10 @@ static const struct rte_pci_id mlx5_pci_id_map[] = {
>  		RTE_PCI_DEVICE(PCI_VENDOR_ID_MELLANOX,
>  				PCI_DEVICE_ID_MELLANOX_BLUEFIELD4)
>  	},
> +	{
> +		RTE_PCI_DEVICE(PCI_VENDOR_ID_MELLANOX,
> +				PCI_DEVICE_ID_MELLANOX_CONNECTX10)
> +	},
>  	{
>  		.vendor_id = 0
>  	}

Similar to the crypto driver, can you verify the indentation here? Does
this code use tabs followed by spaces where only tabs should be used?


More information about the test-report mailing list