https://github.com/torvalds/linux
Revision 916f6efae62305796e012e7c3a7884a267cbacbf authored by Florian Westphal on 17 April 2019, 00:17:23 UTC, committed by Pablo Neira Ayuso on 22 April 2019, 08:34:30 UTC
setting net.netfilter.nf_conntrack_timestamp=1 breaks xmit with fq
scheduler.  skb->tstamp might be "refreshed" using ktime_get_real(),
but fq expects CLOCK_MONOTONIC.

This patch removes all places in netfilter that check/set skb->tstamp:

1. To fix the bogus "start" time seen with conntrack timestamping for
   outgoing packets, never use skb->tstamp and always use current time.
2. In nfqueue and nflog, only use skb->tstamp for incoming packets,
   as determined by current hook (prerouting, input, forward).
3. xt_time has to use system clock as well rather than skb->tstamp.
   We could still use skb->tstamp for prerouting/input/foward, but
   I see no advantage to make this conditional.

Fixes: fb420d5d91c1 ("tcp/fq: move back to CLOCK_MONOTONIC")
Cc: Eric Dumazet <edumazet@google.com>
Reported-by: Michal Soltys <soltys@ziu.info>
Signed-off-by: Florian Westphal <fw@strlen.de>
Acked-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
1 parent 7caa56f
History
Tip revision: 916f6efae62305796e012e7c3a7884a267cbacbf authored by Florian Westphal on 17 April 2019, 00:17:23 UTC
netfilter: never get/set skb->tstamp
Tip revision: 916f6ef
File Mode Size
Makefile -rw-r--r-- 381 bytes
clnt4xdr.c -rw-r--r-- 12.9 KB
clntlock.c -rw-r--r-- 7.4 KB
clntproc.c -rw-r--r-- 21.7 KB
clntxdr.c -rw-r--r-- 13.2 KB
host.c -rw-r--r-- 17.2 KB
mon.c -rw-r--r-- 14.3 KB
netns.h -rw-r--r-- 395 bytes
procfs.c -rw-r--r-- 1.8 KB
procfs.h -rw-r--r-- 474 bytes
svc.c -rw-r--r-- 19.1 KB
svc4proc.c -rw-r--r-- 14.1 KB
svclock.c -rw-r--r-- 25.1 KB
svcproc.c -rw-r--r-- 14.9 KB
svcshare.c -rw-r--r-- 2.4 KB
svcsubs.c -rw-r--r-- 10.3 KB
xdr.c -rw-r--r-- 7.3 KB
xdr4.c -rw-r--r-- 7.6 KB

back to top