2018-05-25 10:14:55

by Yafang Shao

[permalink] [raw]
Subject: [PATCH v2 net-next] tcp: use data length instead of skb->len in tcp_probe

skb->len is meaningless to user.
data length could be more helpful, with which we can easily filter out
the packet without payload.

Signed-off-by: Yafang Shao <[email protected]>
---
include/trace/events/tcp.h | 8 ++++----
1 file changed, 4 insertions(+), 4 deletions(-)

diff --git a/include/trace/events/tcp.h b/include/trace/events/tcp.h
index c1a5284..703abb6 100644
--- a/include/trace/events/tcp.h
+++ b/include/trace/events/tcp.h
@@ -236,7 +236,7 @@
__field(__u16, sport)
__field(__u16, dport)
__field(__u32, mark)
- __field(__u16, length)
+ __field(__u16, data_len)
__field(__u32, snd_nxt)
__field(__u32, snd_una)
__field(__u32, snd_cwnd)
@@ -261,7 +261,7 @@
__entry->dport = ntohs(inet->inet_dport);
__entry->mark = skb->mark;

- __entry->length = skb->len;
+ __entry->data_len = skb->len - tcp_hdrlen(skb);
__entry->snd_nxt = tp->snd_nxt;
__entry->snd_una = tp->snd_una;
__entry->snd_cwnd = tp->snd_cwnd;
@@ -272,9 +272,9 @@
__entry->sock_cookie = sock_gen_cookie(sk);
),

- TP_printk("src=%pISpc dest=%pISpc mark=%#x length=%d snd_nxt=%#x snd_una=%#x snd_cwnd=%u ssthresh=%u snd_wnd=%u srtt=%u rcv_wnd=%u sock_cookie=%llx",
+ TP_printk("src=%pISpc dest=%pISpc mark=%#x data_len=%d snd_nxt=%#x snd_una=%#x snd_cwnd=%u ssthresh=%u snd_wnd=%u srtt=%u rcv_wnd=%u sock_cookie=%llx",
__entry->saddr, __entry->daddr, __entry->mark,
- __entry->length, __entry->snd_nxt, __entry->snd_una,
+ __entry->data_len, __entry->snd_nxt, __entry->snd_una,
__entry->snd_cwnd, __entry->ssthresh, __entry->snd_wnd,
__entry->srtt, __entry->rcv_wnd, __entry->sock_cookie)
);
--
1.8.3.1



2018-05-25 22:03:23

by Song Liu

[permalink] [raw]
Subject: Re: [PATCH v2 net-next] tcp: use data length instead of skb->len in tcp_probe



> On May 25, 2018, at 3:14 AM, Yafang Shao <[email protected]> wrote:
>
> skb->len is meaningless to user.
> data length could be more helpful, with which we can easily filter out
> the packet without payload.
>
> Signed-off-by: Yafang Shao <[email protected]>

Acked-by: Song Liu <[email protected]>


> ---
> include/trace/events/tcp.h | 8 ++++----
> 1 file changed, 4 insertions(+), 4 deletions(-)
>
> diff --git a/include/trace/events/tcp.h b/include/trace/events/tcp.h
> index c1a5284..703abb6 100644
> --- a/include/trace/events/tcp.h
> +++ b/include/trace/events/tcp.h
> @@ -236,7 +236,7 @@
> __field(__u16, sport)
> __field(__u16, dport)
> __field(__u32, mark)
> - __field(__u16, length)
> + __field(__u16, data_len)
> __field(__u32, snd_nxt)
> __field(__u32, snd_una)
> __field(__u32, snd_cwnd)
> @@ -261,7 +261,7 @@
> __entry->dport = ntohs(inet->inet_dport);
> __entry->mark = skb->mark;
>
> - __entry->length = skb->len;
> + __entry->data_len = skb->len - tcp_hdrlen(skb);
> __entry->snd_nxt = tp->snd_nxt;
> __entry->snd_una = tp->snd_una;
> __entry->snd_cwnd = tp->snd_cwnd;
> @@ -272,9 +272,9 @@
> __entry->sock_cookie = sock_gen_cookie(sk);
> ),
>
> - TP_printk("src=%pISpc dest=%pISpc mark=%#x length=%d snd_nxt=%#x snd_una=%#x snd_cwnd=%u ssthresh=%u snd_wnd=%u srtt=%u rcv_wnd=%u sock_cookie=%llx",
> + TP_printk("src=%pISpc dest=%pISpc mark=%#x data_len=%d snd_nxt=%#x snd_una=%#x snd_cwnd=%u ssthresh=%u snd_wnd=%u srtt=%u rcv_wnd=%u sock_cookie=%llx",
> __entry->saddr, __entry->daddr, __entry->mark,
> - __entry->length, __entry->snd_nxt, __entry->snd_una,
> + __entry->data_len, __entry->snd_nxt, __entry->snd_una,
> __entry->snd_cwnd, __entry->ssthresh, __entry->snd_wnd,
> __entry->srtt, __entry->rcv_wnd, __entry->sock_cookie)
> );
> --
> 1.8.3.1
>


2018-05-29 14:18:02

by David Miller

[permalink] [raw]
Subject: Re: [PATCH v2 net-next] tcp: use data length instead of skb->len in tcp_probe

From: Yafang Shao <[email protected]>
Date: Fri, 25 May 2018 18:14:05 +0800

> skb->len is meaningless to user.
> data length could be more helpful, with which we can easily filter out
> the packet without payload.
>
> Signed-off-by: Yafang Shao <[email protected]>

Applied, thank you.

2018-05-29 14:38:30

by Yafang Shao

[permalink] [raw]
Subject: Re: [PATCH v2 net-next] tcp: use data length instead of skb->len in tcp_probe

On Tue, May 29, 2018 at 10:15 PM, David Miller <[email protected]> wrote:
> From: Yafang Shao <[email protected]>
> Date: Fri, 25 May 2018 18:14:05 +0800
>
>> skb->len is meaningless to user.
>> data length could be more helpful, with which we can easily filter out
>> the packet without payload.
>>
>> Signed-off-by: Yafang Shao <[email protected]>
>
> Applied, thank you.

Hi Dave,

There's an update on this patch.
Pls. see V4.

And I will send a V5 patch per Eric's suggestion.

Thanks
Yafang

2018-05-29 14:43:45

by Eric Dumazet

[permalink] [raw]
Subject: Re: [PATCH v2 net-next] tcp: use data length instead of skb->len in tcp_probe



On 05/29/2018 07:36 AM, Yafang Shao wrote:
> On Tue, May 29, 2018 at 10:15 PM, David Miller <[email protected]> wrote:
>> From: Yafang Shao <[email protected]>
>> Date: Fri, 25 May 2018 18:14:05 +0800
>>
>>> skb->len is meaningless to user.
>>> data length could be more helpful, with which we can easily filter out
>>> the packet without payload.
>>>
>>> Signed-off-by: Yafang Shao <[email protected]>
>>
>> Applied, thank you.
>
> Hi Dave,
>
> There's an update on this patch.
> Pls. see V4.
>
> And I will send a V5 patch per Eric's suggestion.

Ah, if it is merged it might be too late.

No big deal.


2018-05-29 14:51:33

by Yafang Shao

[permalink] [raw]
Subject: Re: [PATCH v2 net-next] tcp: use data length instead of skb->len in tcp_probe

On Tue, May 29, 2018 at 10:42 PM, Eric Dumazet <[email protected]> wrote:
>
>
> On 05/29/2018 07:36 AM, Yafang Shao wrote:
>> On Tue, May 29, 2018 at 10:15 PM, David Miller <[email protected]> wrote:
>>> From: Yafang Shao <[email protected]>
>>> Date: Fri, 25 May 2018 18:14:05 +0800
>>>
>>>> skb->len is meaningless to user.
>>>> data length could be more helpful, with which we can easily filter out
>>>> the packet without payload.
>>>>
>>>> Signed-off-by: Yafang Shao <[email protected]>
>>>
>>> Applied, thank you.
>>
>> Hi Dave,
>>
>> There's an update on this patch.
>> Pls. see V4.
>>
>> And I will send a V5 patch per Eric's suggestion.
>
> Ah, if it is merged it might be too late.
>
> No big deal.
>

Since V3, it is changed to two patches. My bad.
Only the original single patch was merged this time.

Will change it based on the new tree.

Thanks
Yafang

2018-05-29 14:56:02

by David Miller

[permalink] [raw]
Subject: Re: [PATCH v2 net-next] tcp: use data length instead of skb->len in tcp_probe

From: Yafang Shao <[email protected]>
Date: Tue, 29 May 2018 22:36:50 +0800

> On Tue, May 29, 2018 at 10:15 PM, David Miller <[email protected]> wrote:
>> From: Yafang Shao <[email protected]>
>> Date: Fri, 25 May 2018 18:14:05 +0800
>>
>>> skb->len is meaningless to user.
>>> data length could be more helpful, with which we can easily filter out
>>> the packet without payload.
>>>
>>> Signed-off-by: Yafang Shao <[email protected]>
>>
>> Applied, thank you.
>
> Hi Dave,
>
> There's an update on this patch.
> Pls. see V4.
>
> And I will send a V5 patch per Eric's suggestion.

Sorry, I pushed it out already. You'll need to send me relative changes.

2018-05-29 14:59:46

by Yafang Shao

[permalink] [raw]
Subject: Re: [PATCH v2 net-next] tcp: use data length instead of skb->len in tcp_probe

On Tue, May 29, 2018 at 10:54 PM, David Miller <[email protected]> wrote:
> From: Yafang Shao <[email protected]>
> Date: Tue, 29 May 2018 22:36:50 +0800
>
>> On Tue, May 29, 2018 at 10:15 PM, David Miller <[email protected]> wrote:
>>> From: Yafang Shao <[email protected]>
>>> Date: Fri, 25 May 2018 18:14:05 +0800
>>>
>>>> skb->len is meaningless to user.
>>>> data length could be more helpful, with which we can easily filter out
>>>> the packet without payload.
>>>>
>>>> Signed-off-by: Yafang Shao <[email protected]>
>>>
>>> Applied, thank you.
>>
>> Hi Dave,
>>
>> There's an update on this patch.
>> Pls. see V4.
>>
>> And I will send a V5 patch per Eric's suggestion.
>
> Sorry, I pushed it out already. You'll need to send me relative changes.

OK