netdev - Re: [PATCH net-next] net/mlx4_en: try to use high order pages for RX rings

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-ID: <d85b28e4-e541-16e7-90e3-06b359bb3dc7@mellanox.com>
Date:   Wed, 7 Mar 2018 11:05:34 +0200
From:   Tariq Toukan <tariqt@...lanox.com>
To:     Eric Dumazet <eric.dumazet@...il.com>,
        David Miller <davem@...emloft.net>
Cc:     netdev <netdev@...r.kernel.org>,
        Tariq Toukan <tariqt@...lanox.com>,
        Eugene Chun <echun@...gle.com>
Subject: Re: [PATCH net-next] net/mlx4_en: try to use high order pages for RX
 rings


On 06/03/2018 9:12 PM, Eric Dumazet wrote:
> From: Eric Dumazet <edumazet@...gle.com>
> 
> RX rings can fit most of the time in a contiguous piece of memory,
> so lets use kvzalloc_node/kvfree instead of vzalloc_node/vfree
> 
> Note that kvzalloc_node() automatically falls back to another node,
> there is no need to do the fallback ourselves.
> 
> Signed-off-by: Eric Dumazet <edumazet@...gle.com>
> ---
>   drivers/net/ethernet/mellanox/mlx4/en_rx.c |   13 +++++--------
>   1 file changed, 5 insertions(+), 8 deletions(-)
> 
> diff --git a/drivers/net/ethernet/mellanox/mlx4/en_rx.c b/drivers/net/ethernet/mellanox/mlx4/en_rx.c
> index c2c6bd7578fdeeee42ded9b10c3993742a244b8f..05787efef492b1c0c6ce540ef73647fad91ce282 100644
> --- a/drivers/net/ethernet/mellanox/mlx4/en_rx.c
> +++ b/drivers/net/ethernet/mellanox/mlx4/en_rx.c
> @@ -291,13 +291,10 @@ int mlx4_en_create_rx_ring(struct mlx4_en_priv *priv,
>   
>   	tmp = size * roundup_pow_of_two(MLX4_EN_MAX_RX_FRAGS *
>   					sizeof(struct mlx4_en_rx_alloc));
> -	ring->rx_info = vzalloc_node(tmp, node);
> +	ring->rx_info = kvzalloc_node(tmp, GFP_KERNEL, node);
>   	if (!ring->rx_info) {
> -		ring->rx_info = vzalloc(tmp);
> -		if (!ring->rx_info) {
> -			err = -ENOMEM;
> -			goto err_xdp_info;
> -		}
> +		err = -ENOMEM;
> +		goto err_xdp_info;
>   	}
>   
>   	en_dbg(DRV, priv, "Allocated rx_info ring at addr:%p size:%d\n",
> @@ -318,7 +315,7 @@ int mlx4_en_create_rx_ring(struct mlx4_en_priv *priv,
>   	return 0;
>   
>   err_info:
> -	vfree(ring->rx_info);
> +	kvfree(ring->rx_info);
>   	ring->rx_info = NULL;
>   err_xdp_info:
>   	xdp_rxq_info_unreg(&ring->xdp_rxq);
> @@ -447,7 +444,7 @@ void mlx4_en_destroy_rx_ring(struct mlx4_en_priv *priv,
>   		bpf_prog_put(old_prog);
>   	xdp_rxq_info_unreg(&ring->xdp_rxq);
>   	mlx4_free_hwq_res(mdev->dev, &ring->wqres, size * stride + TXBB_SIZE);
> -	vfree(ring->rx_info);
> +	kvfree(ring->rx_info);
>   	ring->rx_info = NULL;
>   	kfree(ring);
>   	*pring = NULL;
> 

Reviewed-by: Tariq Toukan <tariqt@...lanox.com>

Thanks Eric.