On Tue, Feb 11, 2025 at 3:12 PM Martin KaFai Lau <martin.lau@xxxxxxxxx> wrote: > > On 2/8/25 2:32 AM, Jason Xing wrote: > > Support SCM_TSTAMP_SCHED case. Introduce SKBTX_BPF used as > > an indicator telling us whether the skb should be traced > > by the bpf prog. > > The BPF side does not exactly support SCM_TSTAMP_SCHED as a report value. > > What this patch does is: > > Add a new sock_ops callback, BPF_SOCK_OPS_TS_SCHED_OPT_CB. This callback will > occur at the same timestamping point as the user space's SCM_TSTAMP_SCHED. The > BPF program can use it to get the same SCM_TSTAMP_SCHED timestamp without > modifying the user-space application. > > A new SKBTX_BPF flag is added to mark skb_shinfo(skb)->tx_flags, ensuring that > the new BPF timestamping and the current user space's SO_TIMESTAMPING do not > interfere with each other. > > I would remove most of the SO_TIMESTAMPING comments from the commit messages. > The timestamping points are the same but there is not much overlapping on the > API side. > > Subject could be: > bpf: Add BPF_SOCK_OPS_TS_SCHED_OPT_CB callback > > [ The same probably for patch 7-9. ] Thanks. I will similarly adjust them as well :) Thanks, Jason
