netdev - Re: [PATCH net 7/8] net/mlx5e: Count GSO packets correctly

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: <CALzJLG_AJASqLheKGv5j=x_Jhp4zfcJdCzhXJYW=Dg7mxvSQ0A@mail.gmail.com>
Date:   Tue, 21 Mar 2017 18:45:54 +0200
From:   Saeed Mahameed <saeedm@....mellanox.co.il>
To:     Eric Dumazet <eric.dumazet@...il.com>
Cc:     Saeed Mahameed <saeedm@...lanox.com>,
        "David S. Miller" <davem@...emloft.net>,
        Linux Netdev List <netdev@...r.kernel.org>,
        Or Gerlitz <ogerlitz@...lanox.com>,
        Gal Pressman <galp@...lanox.com>
Subject: Re: [PATCH net 7/8] net/mlx5e: Count GSO packets correctly

On Tue, Mar 21, 2017 at 6:15 PM, Eric Dumazet <eric.dumazet@...il.com> wrote:
> On Tue, 2017-03-21 at 15:59 +0200, Saeed Mahameed wrote:
>> From: Gal Pressman <galp@...lanox.com>
>>
>> TX packets statistics ('tx_packets' counter) used to count GSO packets
>> as one, even though it contains multiple segments.
>> This patch will increment the counter by the number of segments, and
>> align the driver with the behavior of other drivers in the stack.
>>
>> Note that no information is lost in this patch due to 'tx_tso_packets'
>> counter existence.
>
>> Signed-off-by: Gal Pressman <galp@...lanox.com>
>> Cc: kernel-team@...com
>> Signed-off-by: Saeed Mahameed <saeedm@...lanox.com>
>> ---
>>  drivers/net/ethernet/mellanox/mlx5/core/en_tx.c | 5 +++--
>>  1 file changed, 3 insertions(+), 2 deletions(-)
>>
>> diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_tx.c b/drivers/net/ethernet/mellanox/mlx5/core/en_tx.c
>> index f193128bac4b..57f5e2d7ebd1 100644
>> --- a/drivers/net/ethernet/mellanox/mlx5/core/en_tx.c
>> +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_tx.c
>> @@ -274,15 +274,18 @@ static netdev_tx_t mlx5e_sq_xmit(struct mlx5e_sq *sq, struct sk_buff *skb)
>>                       sq->stats.tso_bytes += skb->len - ihs;
>>               }
>>
>> +             sq->stats.packets += skb_shinfo(skb)->gso_segs;
>>               num_bytes = skb->len + (skb_shinfo(skb)->gso_segs - 1) * ihs;
>
> This reminds me that mlx4 does not trust gso_segs yet, not sure why.

maybe gso_segs wasn't around when mlx4 driver was written :)..

>
>
> diff --git a/drivers/net/ethernet/mellanox/mlx4/en_tx.c b/drivers/net/ethernet/mellanox/mlx4/en_tx.c
> index e0c5ffb3e3a6607456e1f191b0b8c8becfc71219..3ba89bc43d74d8c023776079bcd0bbadd70fb5c6 100644
> --- a/drivers/net/ethernet/mellanox/mlx4/en_tx.c
> +++ b/drivers/net/ethernet/mellanox/mlx4/en_tx.c
> @@ -978,8 +978,7 @@ netdev_tx_t mlx4_en_xmit(struct sk_buff *skb, struct net_device *dev)
>
>                 ring->tso_packets++;
>
> -               i = ((skb->len - lso_header_size) / shinfo->gso_size) +
> -                       !!((skb->len - lso_header_size) % shinfo->gso_size);
> +               i = shinfo->gso_segs;
>                 tx_info->nr_bytes = skb->len + (i - 1) * lso_header_size;
>                 ring->packets += i;
>         } else {
>
>

Right, I Already discussed this with the team, we will fix this

Thank you Eric.