| Message ID | 20230725085443.2102634-1-maze@google.com (mailing list archive) |
|---|---|
| State | Awaiting Upstream |
| Delegated to: | Netdev Maintainers |
| Headers | show |
| Series | [netfilter] netfilter: nfnetlink_log: always add a timestamp | expand |
Maciej Żenczykowski <maze@google.com> wrote: > Compared to all the other work we're already doing to deliver > an skb to userspace this is very cheap - at worse an extra > call to ktime_get_real() - and very useful. Reviewed-by: Florian Westphal <fw@strlen.de>
On Thu, Jul 27, 2023 at 3:58 PM Florian Westphal <fw@strlen.de> wrote: > > Maciej Żenczykowski <maze@google.com> wrote: > > Compared to all the other work we're already doing to deliver > > an skb to userspace this is very cheap - at worse an extra > > call to ktime_get_real() - and very useful. > > Reviewed-by: Florian Westphal <fw@strlen.de> I'm not sure if there's anything else that needs to happen for this to get merged. Maybe it fell through the cracks...? Maybe I didn't add the right CC's... Thanks, Maciej
Maciej Żenczykowski <maze@google.com> wrote: > On Thu, Jul 27, 2023 at 3:58 PM Florian Westphal <fw@strlen.de> wrote: > > > > Maciej Żenczykowski <maze@google.com> wrote: > > > Compared to all the other work we're already doing to deliver > > > an skb to userspace this is very cheap - at worse an extra > > > call to ktime_get_real() - and very useful. > > > > Reviewed-by: Florian Westphal <fw@strlen.de> > > I'm not sure if there's anything else that needs to happen for this to > get merged. This is fine, it'll be merged for -next.
diff --git a/net/netfilter/nfnetlink_log.c b/net/netfilter/nfnetlink_log.c index e57eb168ee13..53c9e76473ba 100644 --- a/net/netfilter/nfnetlink_log.c +++ b/net/netfilter/nfnetlink_log.c @@ -470,7 +470,6 @@ __build_packet_message(struct nfnl_log_net *log, sk_buff_data_t old_tail = inst->skb->tail; struct sock *sk; const unsigned char *hwhdrp; - ktime_t tstamp; nlh = nfnl_msg_put(inst->skb, 0, 0, nfnl_msg_type(NFNL_SUBSYS_ULOG, NFULNL_MSG_PACKET), @@ -599,10 +598,9 @@ __build_packet_message(struct nfnl_log_net *log, goto nla_put_failure; } - tstamp = skb_tstamp_cond(skb, false); - if (hooknum <= NF_INET_FORWARD && tstamp) { + if (hooknum <= NF_INET_FORWARD) { + struct timespec64 kts = ktime_to_timespec64(skb_tstamp_cond(skb, true)); struct nfulnl_msg_packet_timestamp ts; - struct timespec64 kts = ktime_to_timespec64(tstamp); ts.sec = cpu_to_be64(kts.tv_sec); ts.usec = cpu_to_be64(kts.tv_nsec / NSEC_PER_USEC);
Compared to all the other work we're already doing to deliver an skb to userspace this is very cheap - at worse an extra call to ktime_get_real() - and very useful. (and indeed it may even be cheaper if we're running from other hooks) (background: Android occasionally logs packets which caused wake from sleep/suspend and we'd like to have timestamps reliably associated with these events) Cc: Pablo Neira Ayuso <pablo@netfilter.org> Cc: Martin KaFai Lau <kafai@fb.com> Cc: Florian Westphal <fw@strlen.de> Signed-off-by: Maciej Żenczykowski <maze@google.com> --- net/netfilter/nfnetlink_log.c | 6 ++---- 1 file changed, 2 insertions(+), 4 deletions(-)