Re: 2.6.39-rc: WARNING: at __ip_select_ident+0xd3/0xf0()

Le mercredi 23 mars 2011 à 10:14 +0300, Alexander Beregalov a écrit :

Hi

I do not know how to reproduce it.

WARNING: at include/net/inetpeer.h:102 __ip_select_ident+0xd3/0xf0()
Hardware name:
Modules linked in: hwmon_vid sata_sil i2c_nforce2
Pid: 0, comm: swapper Not tainted 2.6.38-07035-g6447f55d #1
Call Trace:
 [<c102ad5d>] warn_slowpath_common+0x6d/0xa0
 [<c13070c3>] ? __ip_select_ident+0xd3/0xf0
 [<c13070c3>] ? __ip_select_ident+0xd3/0xf0
 [<c102adad>] warn_slowpath_null+0x1d/0x20
 [<c13070c3>] __ip_select_ident+0xd3/0xf0
 [<c130df4c>] __ip_make_skb+0x25c/0x360
 [<c130d807>] ? ip_append_data+0x67/0xa0
 [<c132f980>] ? icmp_glue_bits+0x0/0x60
 [<c130e0b4>] ip_push_pending_frames+0x14/0x30
 [<c132f95c>] icmp_push_reply+0xdc/0x100
 [<c13306a8>] icmp_send+0x408/0x4a0
 [<c132ca61>] ? __udp4_lib_lookup.clone.49+0x171/0x2d0
 [<c132d877>] __udp4_lib_rcv+0x487/0x6a0
 [<c13058c0>] ? ip_route_input_common+0x0/0xa00
 [<c132daa2>] udp_rcv+0x12/0x20
 [<c13084bc>] ip_local_deliver+0x9c/0x140
 [<c130846d>] ? ip_local_deliver+0x4d/0x140
 [<c13087c6>] ip_rcv+0x266/0x560
 [<c12e92b3>] __netif_receive_skb+0x393/0x400
 [<c12e8fe4>] ? __netif_receive_skb+0xc4/0x400
 [<c12e9472>] netif_receive_skb+0x42/0x50
 [<c12e94b7>] napi_skb_finish+0x37/0x50
 [<c12eadf3>] napi_gro_receive+0xb3/0xc0
 [<c128d76f>] e1000_receive_skb.clone.34+0x3f/0x60
 [<c128dfcf>] e1000_clean_rx_irq+0x27f/0x4b0
 [<c1055be1>] ? __lock_acquire+0x441/0x19a0
 [<c128d249>] e1000_clean+0x1e9/0x570
 [<c13697fc>] ? _raw_spin_lock+0x5c/0x70
 [<c12e9778>] net_rx_action+0xe8/0x170
 [<c102ffe7>] __do_softirq+0x77/0x110
 [<c102ff70>] ? __do_softirq+0x0/0x110
 <IRQ>  [<c1030246>] ? irq_exit+0x76/0x90
 [<c1003823>] ? do_IRQ+0x43/0xb0
 [<c136f5ee>] ? common_interrupt+0x2e/0x34
 [<c100755f>] ? default_idle+0x2f/0x60
 [<c10015d6>] ? cpu_idle+0x36/0x60
 [<c135f2e5>] ? rest_init+0xa5/0xb0
 [<c135f240>] ? rest_init+0x0/0xb0
 [<c163c663>] ? start_kernel+0x277/0x27e
 [<c163c174>] ? unknown_bootoption+0x0/0x19f
 [<c163c071>] ? i386_start_kernel+0x71/0x77
--

Thanks Alexander for this bug report.

Here is a patch to fix this issue (and other crash you mentioned in next
mail)


[PATCH] ipv4: fix ip_rt_update_pmtu()

commit 2c8cec5c10bc (Cache learned PMTU information in inetpeer) added
an extra inet_putpeer() call in ip_rt_update_pmtu().

This results in various problems, since we can free one inetpeer, while
it is still in use.

Ref: http://www.spinics.net/lists/netdev/msg159121.html

Reported-by: Alexander Beregalov <redacted>
Signed-off-by: Eric Dumazet <redacted>
---
 net/ipv4/route.c |    2 --
 1 file changed, 2 deletions(-)

diff --git a/net/ipv4/route.c b/net/ipv4/route.c
index 870b518..34921b0 100644
--- a/net/ipv4/route.c
+++ b/net/ipv4/route.c

@@ -1593,8 +1593,6 @@ static void ip_rt_update_pmtu(struct dst_entry *dst, u32 mtu)
 			rt->rt_peer_genid = rt_peer_genid();
 		}
 		check_peer_pmtu(dst, peer);
-
-		inet_putpeer(peer);
 	}
 }