[MPTCP] Re: [PATCH v3 3/4] ss: allow dumping MPTCP subflow information

[MPTCP] Re: [PATCH v3 3/4] ss: allow dumping MPTCP subflow information

[Matthieu Baerts]

[-- Attachment #1: Type: text/plain, Size: 1424 bytes --]

Hi Davide,

Thank you for the new version!

On 17/04/2020 18:08, Paolo Abeni wrote:
> From: Davide Caratti <dcaratti(a)redhat.com>
> 
>   [root(a)f31 packetdrill]# ss -tni
> 
>   ESTAB    0        0           192.168.82.247:8080           192.0.2.1:35273
>            cubic wscale:7,8 [...] tcp-ulp-mptcp flags:Mec token:0000(id:0)/5f856c60(id:0) seq:b810457db34209a5 sfseq:1 ssnoff:0 maplen:190
> 
> Signed-off-by: Davide Caratti <dcaratti(a)redhat.com>
> Signed-off-by: Paolo Abeni <pabeni(a)redhat.com>
> ---
>   man/man8/ss.8 |  5 +++++
>   misc/ss.c     | 62 +++++++++++++++++++++++++++++++++++++++++++++++++++
>   2 files changed, 67 insertions(+)
> 
> diff --git a/man/man8/ss.8 b/man/man8/ss.8
> index 023d771b..c80853f9 100644
> --- a/man/man8/ss.8
> +++ b/man/man8/ss.8
> @@ -261,6 +261,11 @@ the pacing rate and max pacing rate
>   .TP
>   .B rcv_space:<rcv_space>
>   a helper variable for TCP internal auto tuning socket receive buffer
> +.P
> +.TP
> +.B tcp-ulp-mptcp flags:[MmBbJjecv] token:<rem_token(rem_id)/loc_token(loc_id)> seq:<sn> sfseq:<ssn> ssnoff:<off> maplen:<maplen>
> +MPTCP subflow information
> +.P

Do you think we can also add what each of "MmBbJjecv" means? :-)

Cheers,
Matt
-- 
Matthieu Baerts | R&D Engineer
matthieu.baerts(a)tessares.net
Tessares SA | Hybrid Access Solutions
www.tessares.net
1 Avenue Jean Monnet, 1348 Louvain-la-Neuve, Belgium

[MPTCP] Re: [PATCH v3 3/4] ss: allow dumping MPTCP subflow information

[Matthieu Baerts]

[-- Attachment #1: Type: text/plain, Size: 1189 bytes --]

Hi Davide,

On 20/04/2020 19:10, Davide Caratti wrote:
> On Mon, 2020-04-20 at 18:47 +0200, Matthieu Baerts wrote:
>> Do you think we can also add what each of "MmBbJjecv" means? :-)
> 
> ... sure, even if it's painful as we are already in a .TP item :) I'm more
> worried about what to write as understandable description for 'sfseq'
> 'ssn_offset', and 'map_valid'.

Indeed, when looking at SS man page, I don't know where to explain that.
Or maybe:

   .B tcp-ulp-mptcp flags:[M<remote MP_CAPABLE>m<local 
MP_CAPABLE>J<remote MP_JOIN>j<local MP_JOIN>B<remote backup>b<local 
backup>e<fully established>c<connected>v<map valid>] 
token:<rem_token(rem_id)/loc_token(loc_id)> seq:<sn> sfseq:<ssn> 
ssnoff:<off> maplen:<maplen>

a bit like what is described for "skmem"?
But that's maybe too long?
Or a link to a doc stored somewhere?

> Can we do this in a separate patch, if it doesn't block the acceptance of
> 'iproute2' series?

Sure, I don't want to block this.

Cheers,
Matt
-- 
Matthieu Baerts | R&D Engineer
matthieu.baerts(a)tessares.net
Tessares SA | Hybrid Access Solutions
www.tessares.net
1 Avenue Jean Monnet, 1348 Louvain-la-Neuve, Belgium

[MPTCP] Re: [PATCH v3 3/4] ss: allow dumping MPTCP subflow information

[Davide Caratti]

[-- Attachment #1: Type: text/plain, Size: 420 bytes --]

On Mon, 2020-04-20 at 18:47 +0200, Matthieu Baerts wrote:
> Do you think we can also add what each of "MmBbJjecv" means? :-)

... sure, even if it's painful as we are already in a .TP item :) I'm more
worried about what to write as understandable description for 'sfseq'
'ssn_offset', and 'map_valid'.

Can we do this in a separate patch, if it doesn't block the acceptance of
'iproute2' series?
-- 
davide