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
File Mode Size
Makefile -rw-r--r-- 1.3 KB
carta_random.S -rw-r--r-- 1.0 KB
checksum.c -rw-r--r-- 2.5 KB
clear_page.S -rw-r--r-- 1.9 KB
clear_user.S -rw-r--r-- 6.1 KB
copy_page.S -rw-r--r-- 2.1 KB
copy_page_mck.S -rw-r--r-- 5.8 KB
copy_user.S -rw-r--r-- 16.7 KB
csum_partial_copy.c -rw-r--r-- 3.0 KB
do_csum.S -rw-r--r-- 10.1 KB
flush.S -rw-r--r-- 2.7 KB
idiv32.S -rw-r--r-- 2.2 KB
idiv64.S -rw-r--r-- 2.2 KB
io.c -rw-r--r-- 2.6 KB
ip_fast_csum.S -rw-r--r-- 2.7 KB
memcpy.S -rw-r--r-- 6.8 KB
memcpy_mck.S -rw-r--r-- 17.4 KB
memset.S -rw-r--r-- 9.1 KB
strlen.S -rw-r--r-- 6.3 KB
strncpy_from_user.S -rw-r--r-- 1.2 KB
strnlen_user.S -rw-r--r-- 996 bytes
xor.S -rw-r--r-- 3.5 KB

back to top