netfilter: nfnetlink: avoid recurrent netns lookups in call_batch
Subsystems (2)
Pablo Neira Ayuso
maintainerPATCH 1/2 netfilter: nfnetlink: avoid recurrent netns lookups in call_batch
Pass the net pointer to the call_batch callback functions so we can skip recurrent lookups.
---
I would really appreciate to get a Tested-by: tag from you on this.
include/linux/netfilter/nfnetlink.h | 2 +- net/netfilter/nf_tables_api.c | 96 +++++++++++++++++-------------------- net/netfilter/nfnetlink.c | 2 +- 3 files changed, 47 insertions(+), 53 deletions(-)
include/linux/netfilter/nfnetlink.h +1 -1
@@ -14,7 +14,7 @@ struct nfnl_callback {int (*call_rcu)(struct sock *nl,struct sk_buff *skb,const struct nlmsghdr *nlh,const struct nlattr *const cda[]); -int (*call_batch)(struct sock *nl,struct sk_buff *skb, +int (*call_batch)(struct net *net,struct sock *nl,struct sk_buff *skb,const struct nlmsghdr *nlh,const struct nlattr *const cda[]);const struct nla_policy *policy;/* netlink attribute policy */
net/netfilter/nf_tables_api.c +45 -51
343 lines not on this page · raw message
net/netfilter/nfnetlink.c +1 -1
@@ -381,7 +381,7 @@ replay: goto ack;if (nc->call_batch) { - err = nc->call_batch (net->nfnl, skb, nlh, + err = nc->call_batch (net, net->nfnl, skb, nlh, (const struct nlattr **)cda); }
Signature · 1 line
2.1.4
Pablo Neira Ayuso
maintainerPATCH 2/2 nfnetlink: fix splat due to incorrect socket memory accounting in skbuff clones
If we attach the sk to the skb, netlink_skb_destructor() will underflow the socket receive memory counter and we get warning splat when releasing the socket.
$ cat /proc/net/netlink
sk Eth Pid Groups Rmem Wmem Dump Locks Drops Inode
ffff8800ca903000 12 0 00000000 -54144 0 0 2 0 17942
^^^^^^Rmem above shows an underflow.
And here below the warning splat:
[ 1363.815976] WARNING: CPU: 2 PID: 1356 at net/netlink/af_netlink.c:958 netlink_sock_destruct+0x80/0xb9() [...] [ 1363.816152] CPU: 2 PID: 1356 Comm: kworker/u16:1 Tainted: G W 4.4.0-rc1+ #153 [ 1363.816155] Hardware name: LENOVO 23259H1/23259H1, BIOS G2ET32WW (1.12 ) 05/30/2012 [ 1363.816160] Workqueue: netns cleanup_net [ 1363.816163] 0000000000000000 ffff880119203dd0 ffffffff81240204 0000000000000000 [ 1363.816169] ffff880119203e08 ffffffff8104db4b ffffffff813d49a1 ffff8800ca771000 [ 1363.816174] ffffffff81a42b00 0000000000000000 ffff8800c0afe1e0 ffff880119203e18 [ 1363.816179] Call Trace: [ 1363.816181] <IRQ> [<ffffffff81240204>] dump_stack+0x4e/0x79 [ 1363.816193] [<ffffffff8104db4b>] warn_slowpath_common+0x9a/0xb3 [ 1363.816197] [<ffffffff813d49a1>] ? netlink_sock_destruct+0x80/0xb9
skb->sk was only needed to lookup for the netns, however we don't need this anymore since ("netfilter: nfnetlink: avoid recurrent netns lookups in call_batch"), so this patch removes this manual socket assignment.
---
I would really appreciate to get a Tested-by: tag from you on this.
net/netfilter/nfnetlink.c | 2 -- 1 file changed, 2 deletions(-)
net/netfilter/nfnetlink.c +0 -2
@@ -295,8 +295,6 @@ replay:if (!skb)return netlink_ack (oskb, nlh, -ENOMEM ); - skb->sk = oskb->sk; -nfnl_lock (subsys_id); ss =rcu_dereference_protected (table[subsys_id].subsys,lockdep_is_held (&table[subsys_id].mutex));
Signature · 1 line
2.1.4
Re: [PATCH nf 2/2] nfnetlink: fix splat due to incorrect socket memory accounting in skbuff clones
Quoted 30 lines · Pablo Neira Ayuso
If we attach the sk to the skb,
netlink_skb_destructor()will underflow the socket receive memory counter and we get warning splat when releasing the socket.$ cat /proc/net/netlink sk Eth Pid Groups Rmem Wmem Dump Locks Drops Inode ffff8800ca903000 12 0 00000000 -54144 0 0 2 0 17942 ^^^^^^Rmem above shows an underflow.
And here below the warning splat:
[ 1363.815976] WARNING: CPU: 2 PID: 1356 at net/netlink/af_netlink.c:958 netlink_sock_destruct+0x80/0xb9() [...] [ 1363.816152] CPU: 2 PID: 1356 Comm: kworker/u16:1 Tainted: G W 4.4.0-rc1+ #153 [ 1363.816155] Hardware name: LENOVO 23259H1/23259H1, BIOS G2ET32WW (1.12 ) 05/30/2012 [ 1363.816160] Workqueue: netns cleanup_net [ 1363.816163] 0000000000000000 ffff880119203dd0 ffffffff81240204 0000000000000000 [ 1363.816169] ffff880119203e08 ffffffff8104db4b ffffffff813d49a1 ffff8800ca771000 [ 1363.816174] ffffffff81a42b00 0000000000000000 ffff8800c0afe1e0 ffff880119203e18 [ 1363.816179] Call Trace: [ 1363.816181] <IRQ> [<ffffffff81240204>] dump_stack+0x4e/0x79 [ 1363.816193] [<ffffffff8104db4b>] warn_slowpath_common+0x9a/0xb3 [ 1363.816197] [<ffffffff813d49a1>] ? netlink_sock_destruct+0x80/0xb9
skb->skwas only needed to lookup for the netns, however we don't need this anymore since ("netfilter: nfnetlink: avoid recurrent netns lookups incall_batch"), so this patch removes this manual socket assignment.
Quoted 8 lines
Reported-by Arturo Borrero Gonzalez <[email protected]>Reported-by Ben Hutchings <[email protected]>Signed-off-by Pablo Neira Ayuso <[email protected]>---
I would really appreciate to get a Tested-by: tag from you on this.net/netfilter/nfnetlink.c | 2 -- 1 file changed, 2 deletions(-)
thanks, the problem seems to be fixed now.
Signature · 5 lines
Arturo Borrero González -- To unsubscribe from this list: send the line "unsubscribe netfilter-devel" in the body of a message to [email protected] More majordomo info at http://vger.kernel.org/majordomo-info.html
Pablo Neira Ayuso
maintainerRe: [PATCH nf 2/2] nfnetlink: fix splat due to incorrect socket memory accounting in skbuff clones
Quoted 37 lines · Arturo Borrero Gonzalez
Quoted 36 lines · Pablo Neira Ayuso
If we attach the sk to the skb,
netlink_skb_destructor()will underflow the socket receive memory counter and we get warning splat when releasing the socket.$ cat /proc/net/netlink sk Eth Pid Groups Rmem Wmem Dump Locks Drops Inode ffff8800ca903000 12 0 00000000 -54144 0 0 2 0 17942 ^^^^^^Rmem above shows an underflow.
And here below the warning splat:
[ 1363.815976] WARNING: CPU: 2 PID: 1356 at net/netlink/af_netlink.c:958 netlink_sock_destruct+0x80/0xb9() [...] [ 1363.816152] CPU: 2 PID: 1356 Comm: kworker/u16:1 Tainted: G W 4.4.0-rc1+ #153 [ 1363.816155] Hardware name: LENOVO 23259H1/23259H1, BIOS G2ET32WW (1.12 ) 05/30/2012 [ 1363.816160] Workqueue: netns cleanup_net [ 1363.816163] 0000000000000000 ffff880119203dd0 ffffffff81240204 0000000000000000 [ 1363.816169] ffff880119203e08 ffffffff8104db4b ffffffff813d49a1 ffff8800ca771000 [ 1363.816174] ffffffff81a42b00 0000000000000000 ffff8800c0afe1e0 ffff880119203e18 [ 1363.816179] Call Trace: [ 1363.816181] <IRQ> [<ffffffff81240204>] dump_stack+0x4e/0x79 [ 1363.816193] [<ffffffff8104db4b>] warn_slowpath_common+0x9a/0xb3 [ 1363.816197] [<ffffffff813d49a1>] ? netlink_sock_destruct+0x80/0xb9
skb->skwas only needed to lookup for the netns, however we don't need this anymore since ("netfilter: nfnetlink: avoid recurrent netns lookups incall_batch"), so this patch removes this manual socket assignment.Reported-by Arturo Borrero Gonzalez <[email protected]>Reported-by Ben Hutchings <[email protected]>Signed-off-by Pablo Neira Ayuso <[email protected]>---
I would really appreciate to get a Tested-by: tag from you on this.
Quoted 6 lines
Quoted 2 lines
net/netfilter/nfnetlink.c | 2 -- 1 file changed, 2 deletions(-)thanks, the problem seems to be fixed now.
Thanks for testing Arturo!
It would be good to give a another testing given this is related to netns as well: http://patchwork.ozlabs.org/patch/554791/. What I could test here showed no problems.
Quoted 12 lines · Pablo Neira Ayuso
Pass the net pointer to the
call_batchcallback functions so we can skip recurrent lookups.Signed-off-by Pablo Neira Ayuso <[email protected]>---
I would really appreciate to get a Tested-by: tag from you on this.include/linux/netfilter/nfnetlink.h | 2 +- net/netfilter/nf_tables_api.c | 96 +++++++++++++++++-------------------- net/netfilter/nfnetlink.c | 2 +- 3 files changed, 47 insertions(+), 53 deletions(-)
thanks, the problem seems to be fixed now.
Signature · 5 lines
Arturo Borrero González -- To unsubscribe from this list: send the line "unsubscribe netfilter-devel" in the body of a message to [email protected] More majordomo info at http://vger.kernel.org/majordomo-info.html