On Thu, 2017-03-09 at 15:27 +0200, Sagi Grimberg wrote:
> Hey all,
>
> I get this with nfs mount on my VM (4.11-rc1):
>
> from dmesg:
> [ 1889.676648] kmemleak: 8254 new suspected memory leaks (see
> /sys/kernel/debug/kmemleak)
>
> $ cat /sys/kernel/debug/kmemleak
> ...
> unreferenced object 0xffff96c67a192a80 (size 32):
> comm "kworker/0:3", pid 266, jiffies 4295080946 (age 6406.600s)
> hex dump (first 32 bytes):
> 31 30 30 30 00 ff 0f 88 0a fc ff ff 41 83 ff
> 01 1000........A...
> 0f 8e 75 01 00 00 41 8d 4f fc 83 f9 0b 0f 87
> f3 ..u...A.O.......
> backtrace:
> [<ffffffffbb847eda>] kmemleak_alloc+0x4a/0xa0
> [<ffffffffbb205949>] __kmalloc+0x149/0x270
> [<ffffffffc0526590>] xdr_stream_decode_string_dup+0x80/0x110
> [sunrpc]
> [<ffffffffc0a03168>] decode_getfattr_attrs+0x8b8/0x15a0 [nfsv4]
> [<ffffffffc0a03edc>]
> decode_getfattr_generic.constprop.108+0x8c/0xe0 [nfsv4]
> [<ffffffffc0a03fdd>] nfs4_xdr_dec_open+0xad/0xd0 [nfsv4]
> [<ffffffffc051cf99>] rpcauth_unwrap_resp+0x89/0xd0 [sunrpc]
> [<ffffffffc050db73>] call_decode+0x1e3/0x810 [sunrpc]
> [<ffffffffc0519632>] __rpc_execute+0x82/0x450 [sunrpc]
> [<ffffffffc0519a12>] rpc_async_schedule+0x12/0x20 [sunrpc]
> [<ffffffffbb099e3b>] process_one_work+0x16b/0x480
> [<ffffffffbb09a19b>] worker_thread+0x4b/0x500
> [<ffffffffbb0a0591>] kthread+0x101/0x140
> [<ffffffffbb85303c>] ret_from_fork+0x2c/0x40
> [<ffffffffffffffff>] 0xffffffffffffffff
> unreferenced object 0xffff96c67a192a60 (size 32):
> comm "kworker/0:3", pid 266, jiffies 4295080947 (age 6406.596s)
> hex dump (first 32 bytes):
> 31 30 30 30 00 ff ff 41 c6 85 90 00 00 00 00
> 41 1000...A.......A
> 0f b6 85 91 00 00 00 88 45 bf e9 db fd ff ff
> 44 ........E......D
> backtrace:
> [<ffffffffbb847eda>] kmemleak_alloc+0x4a/0xa0
> [<ffffffffbb205949>] __kmalloc+0x149/0x270
> [<ffffffffc0526590>] xdr_stream_decode_string_dup+0x80/0x110
> [sunrpc]
> [<ffffffffc0a03168>] decode_getfattr_attrs+0x8b8/0x15a0 [nfsv4]
> [<ffffffffc0a03edc>]
> decode_getfattr_generic.constprop.108+0x8c/0xe0 [nfsv4]
> [<ffffffffc0a03fdd>] nfs4_xdr_dec_open+0xad/0xd0 [nfsv4]
> [<ffffffffc051cf99>] rpcauth_unwrap_resp+0x89/0xd0 [sunrpc]
> [<ffffffffc050db73>] call_decode+0x1e3/0x810 [sunrpc]
> [<ffffffffc0519632>] __rpc_execute+0x82/0x450 [sunrpc]
> [<ffffffffc0519a12>] rpc_async_schedule+0x12/0x20 [sunrpc]
> [<ffffffffbb099e3b>] process_one_work+0x16b/0x480
> [<ffffffffbb09a19b>] worker_thread+0x4b/0x500
> [<ffffffffbb0a0591>] kthread+0x101/0x140
> [<ffffffffbb85303c>] ret_from_fork+0x2c/0x40
> [<ffffffffffffffff>] 0xffffffffffffffff
That's a known issue due to a typo in the getattr cleanups. See the
patch posted by Kinglong to this mailing list a couple of days ago. I'm
sure Anna is planning on merging it soon.
--
Trond Myklebust
Linux NFS client maintainer, PrimaryData
trond.myklebust@xxxxxxxxxxxxxxx
��.n��������+%������w��{.n�����{��w��� jg��� �����ݢj����G���� ���j:+v���w�m������w�������h�����٥
