Re: [PATCH v1 net-next] ila: avoid genlmsg_reply when not ila_map found

[Simon Horman]

On Wed, May 22, 2024 at 11:15:37AM +0800, Lin Ma wrote:
> The current ila_xlat_nl_cmd_get_mapping will call genlmsg_reply even if
> not ila_map found with user provided parameters. Then an empty netlink
> message will be sent and cause a WARNING like below.
> 
> WARNING: CPU: 1 PID: 7709 at include/linux/skbuff.h:2524 __dev_queue_xmit+0x241b/0x3b60 net/core/dev.c:4171
> Modules linked in:
> Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS 1.13.0-1ubuntu1.1 04/01/2014
> RIP: 0010:skb_assert_len include/linux/skbuff.h:2524 [inline]
> RIP: 0010:__dev_queue_xmit+0x241b/0x3b60 net/core/dev.c:4171
> RSP: 0018:ffffc9000f90f228 EFLAGS: 00010282
> RAX: 0000000000000000 RBX: 0000000000000000 RCX: 0000000000000000
> RDX: 0000000000040000 RSI: ffffffff816968a8 RDI: fffff52001f21e37
> RBP: ffff8881077f2d3a R08: 0000000000000005 R09: 0000000000000000
> R10: 0000000080000000 R11: 0000000000000000 R12: dffffc0000000000
> R13: 0000000000000000 R14: ffff8881077f2c90 R15: ffff8881077f2c80
> FS:  00007fb8be338700(0000) GS:ffff888135c00000(0000) knlGS:0000000000000000
> CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
> CR2: 000000c00ca8a000 CR3: 0000000105e17000 CR4: 0000000000150ee0
> Call Trace:
>  <TASK>
>  dev_queue_xmit include/linux/netdevice.h:3008 [inline]
>  __netlink_deliver_tap_skb net/netlink/af_netlink.c:307 [inline]
>  __netlink_deliver_tap net/netlink/af_netlink.c:325 [inline]
>  netlink_deliver_tap+0x9e4/0xc40 net/netlink/af_netlink.c:338
>  __netlink_sendskb net/netlink/af_netlink.c:1263 [inline]
>  netlink_sendskb net/netlink/af_netlink.c:1272 [inline]
>  netlink_unicast+0x6ac/0x7f0 net/netlink/af_netlink.c:1360
>  nlmsg_unicast include/net/netlink.h:1067 [inline]
>  genlmsg_unicast include/net/genetlink.h:372 [inline]
>  genlmsg_reply include/net/genetlink.h:382 [inline]
>  ila_xlat_nl_cmd_get_mapping+0x589/0x950 net/ipv6/ila/ila_xlat.c:493
>  genl_family_rcv_msg_doit+0x228/0x320 net/netlink/genetlink.c:756
>  genl_family_rcv_msg net/netlink/genetlink.c:833 [inline]
>  genl_rcv_msg+0x441/0x780 net/netlink/genetlink.c:850
>  netlink_rcv_skb+0x153/0x400 net/netlink/af_netlink.c:2540
>  genl_rcv+0x24/0x40 net/netlink/genetlink.c:861
>  netlink_unicast_kernel net/netlink/af_netlink.c:1319 [inline]
>  netlink_unicast+0x543/0x7f0 net/netlink/af_netlink.c:1345
>  netlink_sendmsg+0x917/0xe10 net/netlink/af_netlink.c:1921
>  sock_sendmsg_nosec net/socket.c:714 [inline]
>  sock_sendmsg+0xcf/0x120 net/socket.c:734
>  ____sys_sendmsg+0x6eb/0x810 net/socket.c:2482
>  ___sys_sendmsg+0x110/0x1b0 net/socket.c:2536
>  __sys_sendmsg+0xf3/0x1c0 net/socket.c:2565
>  do_syscall_x64 arch/x86/entry/common.c:50 [inline]
>  do_syscall_64+0x35/0xb0 arch/x86/entry/common.c:80
>  entry_SYSCALL_64_after_hwframe+0x63/0xcd
> RIP: 0033:0x7fb8bd68f359
> Code: 28 00 00 00 75 05 48 83 c4 28 c3 e8 f1 19 00 00 90 48 89 f8 48 89 f7 48 89 d6 48 89 ca 4d 89 c2 4d 89 c8 4c 8b 4c 24 08 0f 05 <48> 3d 01 f0 ff ff 73 01 c3 48 c7 c1 b8 ff ff ff f7 d8 64 89 01 48
> RSP: 002b:00007fb8be338168 EFLAGS: 00000246 ORIG_RAX: 000000000000002e
> RAX: ffffffffffffffda RBX: 00007fb8bd7bbf80 RCX: 00007fb8bd68f359
> RDX: 0000000000000000 RSI: 0000000020000100 RDI: 0000000000000005
> RBP: 00007fb8bd6da498 R08: 0000000000000000 R09: 0000000000000000
> R10: 0000000000000000 R11: 0000000000000246 R12: 0000000000000000
> R13: 00007ffc22fb52af R14: 00007fb8be338300 R15: 0000000000022000
> 
> Do nullptr check and assign -EINVAL ret if no ila_map found.
> 
> Signed-off-by: Lin Ma <linma@xxxxxxxxxx>
> ---
>  net/ipv6/ila/ila_xlat.c | 2 ++
>  1 file changed, 2 insertions(+)
> 
> diff --git a/net/ipv6/ila/ila_xlat.c b/net/ipv6/ila/ila_xlat.c
> index 67e8c9440977..63b8fe1b8255 100644
> --- a/net/ipv6/ila/ila_xlat.c
> +++ b/net/ipv6/ila/ila_xlat.c

Hi Lin Ma,

The lines immediately above those covered by this patch are as follows:

	ret = -ESRCH;
	ila = ila_lookup_by_params(&xp, ilan);
	if (ila) {
		ret = ila_dump_info(ila,

> @@ -483,6 +483,8 @@ int ila_xlat_nl_cmd_get_mapping(struct sk_buff *skb, struct genl_info *info)
>  				    info->snd_portid,
>  				    info->snd_seq, 0, msg,
>  				    info->genlhdr->cmd);
> +	} else {
> +		ret = -EINVAL;
>  	}
>  
>  	rcu_read_unlock();

And the lines following, up to the end of the function, are:

	if (ret < 0)
		goto out_free;

	return genlmsg_reply(msg, info);

out_free:
	nlmsg_free(msg);
	return ret;

By my reading, without your patch, if ila is not found (NULL)
then ret will be -ESRCH, and genlmsg_reply will not be called.

I feel that I am missing something here.