netdev - Re: [RFC PATCH net-next v6 13/13] bpf: add simple bpf tests in the tx path for so

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: <29073a9e-23ea-49c2-b0ad-d33bd3ea8974@linux.dev>
Date: Mon, 27 Jan 2025 15:49:11 -0800
From: Martin KaFai Lau <martin.lau@...ux.dev>
To: Jason Xing <kerneljasonxing@...il.com>
Cc: davem@...emloft.net, edumazet@...gle.com, kuba@...nel.org,
 pabeni@...hat.com, dsahern@...nel.org, willemdebruijn.kernel@...il.com,
 willemb@...gle.com, ast@...nel.org, daniel@...earbox.net, andrii@...nel.org,
 eddyz87@...il.com, song@...nel.org, yonghong.song@...ux.dev,
 john.fastabend@...il.com, kpsingh@...nel.org, sdf@...ichev.me,
 haoluo@...gle.com, jolsa@...nel.org, horms@...nel.org, bpf@...r.kernel.org,
 netdev@...r.kernel.org
Subject: Re: [RFC PATCH net-next v6 13/13] bpf: add simple bpf tests in the tx
 path for so_timestamping feature

On 1/24/25 7:42 PM, Jason Xing wrote:
>> Please also add some details on how the UDP BPF_SOCK_OPS_TS_TCP_SND_CB (or to be
>> renamed to BPF_SOCK_OPS_TS_SND_CB ?) will look like. It is the only callback
>> that I don't have a clear idea for UDP.
> I think I will rename it as you said. But I wonder if I can add more
> details about UDP after this series gets merged which should not be
> too late. After this series, I will carefully consider and test how we
> use for UDP type.

Not asking for a full UDP implementation, having this set staying with TCP is 
ok. We have pretty clear idea on all the new TS_*_CB will work in UDP except the 
TS_SND_CB.

I am asking at least a description on where this SND hook will be in UDP and how 
the delay will be measured from the udp_sendmsg(). I haven't looked, so the 
question. It is better to get some visibility first instead of scrambling to 
change it after landing to -next.