Re: [PATCH bpf] bpf: check for data_len before upgrading mss when 6 to 4

(off-list ancestor, not in this archive)
[PATCH bpf] bpf: check for data_len before upgrading mss when 6 to 4 · Dongseok Yi <hidden> · 2021-04-29
Re: [PATCH bpf] bpf: check for data_len before upgrading mss when 6 to 4 · Daniel Borkmann <daniel@iogearbox.net> · 2021-05-05
RE: [PATCH bpf] bpf: check for data_len before upgrading mss when 6 to 4 · Dongseok Yi <hidden> · 2021-05-06
Re: [PATCH bpf] bpf: check for data_len before upgrading mss when 6 to 4 · Willem de Bruijn <willemdebruijn.kernel@gmail.com> · 2021-05-06
RE: [PATCH bpf] bpf: check for data_len before upgrading mss when 6 to 4 · Dongseok Yi <hidden> · 2021-05-06
Re: [PATCH bpf] bpf: check for data_len before upgrading mss when 6 to 4 · Willem de Bruijn <willemdebruijn.kernel@gmail.com> · 2021-05-06
RE: [PATCH bpf] bpf: check for data_len before upgrading mss when 6 to 4 · Dongseok Yi <hidden> · 2021-05-07
Re: [PATCH bpf] bpf: check for data_len before upgrading mss when 6 to 4 · Willem de Bruijn <willemdebruijn.kernel@gmail.com> · 2021-05-07
Re: [PATCH bpf] bpf: check for data_len before upgrading mss when 6 to 4 · Yunsheng Lin <hidden> · 2021-05-07
Re: [PATCH bpf] bpf: check for data_len before upgrading mss when 6 to 4 · Willem de Bruijn <willemdebruijn.kernel@gmail.com> · 2021-05-07
RE: [PATCH bpf] bpf: check for data_len before upgrading mss when 6 to 4 · Dongseok Yi <hidden> · 2021-05-07
Re: [PATCH bpf] bpf: check for data_len before upgrading mss when 6 to 4 · Yunsheng Lin <hidden> · 2021-05-07
RE: [PATCH bpf] bpf: check for data_len before upgrading mss when 6 to 4 · Dongseok Yi <hidden> · 2021-05-07
Re: [PATCH bpf] bpf: check for data_len before upgrading mss when 6 to 4 · Willem de Bruijn <willemdebruijn.kernel@gmail.com> · 2021-05-07
RE: [PATCH bpf] bpf: check for data_len before upgrading mss when 6 to 4 · Dongseok Yi <hidden> · 2021-05-10
Re: [PATCH bpf] bpf: check for data_len before upgrading mss when 6 to 4 · Willem de Bruijn <willemdebruijn.kernel@gmail.com> · 2021-05-10
Re: [PATCH bpf] bpf: check for data_len before upgrading mss when 6 to 4 · Willem de Bruijn <willemdebruijn.kernel@gmail.com> · 2021-05-10
RE: [PATCH bpf] bpf: check for data_len before upgrading mss when 6 to 4 · Dongseok Yi <hidden> · 2021-05-11
Re: [PATCH bpf] bpf: check for data_len before upgrading mss when 6 to 4 · Willem de Bruijn <willemdebruijn.kernel@gmail.com> · 2021-05-11
RE: [PATCH bpf] bpf: check for data_len before upgrading mss when 6 to 4 · Dongseok Yi <hidden> · 2021-05-12
[PATCH bpf v2] bpf: check BPF_F_ADJ_ROOM_FIXED_GSO when upgrading mss in 6 to 4 · Dongseok Yi <hidden> · 2021-05-11
Re: [PATCH bpf v2] bpf: check BPF_F_ADJ_ROOM_FIXED_GSO when upgrading mss in 6 to 4 · Willem de Bruijn <willemdebruijn.kernel@gmail.com> · 2021-05-11
RE: [PATCH bpf v2] bpf: check BPF_F_ADJ_ROOM_FIXED_GSO when upgrading mss in 6 to 4 · Dongseok Yi <hidden> · 2021-05-12
[PATCH bpf-next v3] bpf: check for BPF_F_ADJ_ROOM_FIXED_GSO when bpf_skb_change_proto · Dongseok Yi <hidden> · 2021-05-12
Re: [PATCH bpf-next v3] bpf: check for BPF_F_ADJ_ROOM_FIXED_GSO when bpf_skb_change_proto · Willem de Bruijn <willemdebruijn.kernel@gmail.com> · 2021-05-12
Re: [PATCH bpf-next v3] bpf: check for BPF_F_ADJ_ROOM_FIXED_GSO when bpf_skb_change_proto · patchwork-bot+netdevbpf@kernel.org · 2021-05-18

From: Willem de Bruijn <willemdebruijn.kernel@gmail.com>
Date: 2021-05-10 13:50:01
Also in: bpf, lkml

quoted

That generates TCP packets with different MSS within the same stream.

My suggestion remains to just not change MSS at all. But this has to
be a new flag to avoid changing established behavior.

I don't understand why the mss size should be kept in GSO step. Will
there be any issue with different mss?

This issue has come up before and that has been the feedback from
TCP experts at one point.

In general, upgrading mss make sense when 6 to 4. The new flag would be
set by user to not change mss. What happened if user does not set the
flag? I still think we should fix the issue with a general approach. Or
can we remove the skb_increase_gso_size line?

Admins that insert such BPF packets should be aware of these issues.
And likely be using clamping. This is a known issue.

We arrived that the flag approach in bpf_skb_net_shrink. Extending
that  to bpf_skb_change_proto would be consistent.

`h`	back out one level
`j`	next message in thread
`k`	previous message in thread
`l`	drill in
`Esc`	close help / fold thread tree
`?`	toggle this help