[<prev] [next>] [<thread-prev] [day] [month] [year] [list]
Message-ID: <2df8d7f1-2c62-7d72-aae4-4c1276b2857f@seti.kr.ua>
Date: Tue, 21 Apr 2020 21:27:40 +0300
From: Andrew <nitr0@...i.kr.ua>
To: netdev@...r.kernel.org
Subject: Re: Memory leaks in 4.14.160+
Maybe some extra info is needed?
19.04.2020 12:24, Andrew пишет:
>
>
> Hi all.
>
> I noticed kernel memory leakage in 4.14.160 kernel (~1GB per month on
> my setup). kmemleak reported a lot of small leaks in 2 places:
>
> unreferenced object 0xffff88806f0dbe00 (size 256):
> comm "softirq", pid 0, jiffies 8895939313 (age 422075.624s)
> hex dump (first 32 bytes):
> 01 00 00 00 00 00 00 00 02 00 00 00 00 03 00 00 ................
> c9 ff 0a 00 00 00 00 00 20 ff 6b e7 80 88 ff ff ........ .k.....
> backtrace:
> [<ffffffff816cf9e8>] kmemleak_alloc+0x28/0x50
> [<ffffffff8119d405>] kmem_cache_alloc+0x125/0x190
> [<ffffffffa01b3183>] __nf_conntrack_alloc+0x53/0x190 [nf_conntrack]
> [<ffffffffa01b3786>] init_conntrack+0x4a6/0x500 [nf_conntrack]
> [<ffffffffa01b3bf6>] nf_conntrack_in+0x416/0x470 [nf_conntrack]
> [<ffffffffa016124c>] ipv4_conntrack_in+0x1c/0x20 [nf_conntrack_ipv4]
> [<ffffffff816338b8>] nf_hook_slow+0x48/0xd0
> [<ffffffff8163c49f>] ip_rcv+0x33f/0x380
> [<ffffffff815ede07>] __netif_receive_skb_core+0x8a7/0xc20
> [<ffffffff815f0828>] __netif_receive_skb+0x18/0x60
> [<ffffffff815f0dd1>] process_backlog+0x91/0x150
> [<ffffffff815f448b>] net_rx_action+0x2bb/0x780
> [<ffffffff81a0010c>] __do_softirq+0x10c/0x2b0
> [<ffffffff8106f754>] irq_exit+0xc4/0xd0
> [<ffffffff8180266f>] do_IRQ+0x4f/0xe0
> [<ffffffff8180094c>] ret_from_intr+0x0/0x1d
> unreferenced object 0xffff8880c3c1c180 (size 128):
> comm "softirq", pid 0, jiffies 8895939313 (age 422075.624s)
> hex dump (first 32 bytes):
> 82 00 63 3b ce fb 6a 8a 00 00 00 00 00 00 00 00 ..c;..j.........
> 00 00 00 00 20 00 00 00 00 38 c1 c3 80 88 ff ff .... ....8......
> backtrace:
> [<ffffffff816cf9e8>] kmemleak_alloc+0x28/0x50
> [<ffffffff8119fc5c>] __kmalloc_track_caller+0x13c/0x1b0
> [<ffffffff8116b141>] __krealloc+0x51/0x90
> [<ffffffffa01baf37>] nf_ct_ext_add+0x97/0x180 [nf_conntrack]
> [<ffffffffa01b34be>] init_conntrack+0x1de/0x500 [nf_conntrack]
> [<ffffffffa01b3bf6>] nf_conntrack_in+0x416/0x470 [nf_conntrack]
> [<ffffffffa016124c>] ipv4_conntrack_in+0x1c/0x20 [nf_conntrack_ipv4]
> [<ffffffff816338b8>] nf_hook_slow+0x48/0xd0
> [<ffffffff8163c49f>] ip_rcv+0x33f/0x380
> [<ffffffff815ede07>] __netif_receive_skb_core+0x8a7/0xc20
> [<ffffffff815f0828>] __netif_receive_skb+0x18/0x60
> [<ffffffff815f0dd1>] process_backlog+0x91/0x150
> [<ffffffff815f448b>] net_rx_action+0x2bb/0x780
> [<ffffffff81a0010c>] __do_softirq+0x10c/0x2b0
> [<ffffffff8106f754>] irq_exit+0xc4/0xd0
> [<ffffffff8180266f>] do_IRQ+0x4f/0xe0
>
> 4.14.174 have same leaks.
>
> System is used as NAT box + PPPoE terminator, conntrack helpers are
> loaded automatically by net.netfilter.nf_conntrack_helper=1; loaded
> helpers: nf_nat_sip nf_nat_pptp nf_nat_ftp
>
>
Powered by blists - more mailing lists