[bug report] kmemleak observed with blktests nvme/tcp
Yi Zhang
yi.zhang at redhat.com
Thu Apr 18 18:04:20 PDT 2024
On Tue, Apr 16, 2024 at 11:19 AM Chaitanya Kulkarni
<chaitanyak at nvidia.com> wrote:
>
> +linux-nvme list for awareness ...
>
> -ck
>
>
> On 4/6/24 17:38, Yi Zhang wrote:
> > Hello
> >
> > I found the kmemleak issue after blktests nvme/tcp tests on the latest
> > linux-block/for-next, please help check it and let me know if you need
> > any info/testing for it, thanks.
>
> it will help others to specify which testcase you are using ...
I finally figured out it was triggered by nvme/012, and after the
test, I also found some I/O timeout from dmesg, here is the log:
# nvme_trtype=tcp ./check nvme/012
nvme/012 (run mkfs and data verification fio job on NVMeOF block
device-backed ns) [passed]
runtime 303.431s ... 384.831s
[40886.926741] loop: module loaded
[40887.798283] run blktests nvme/012 at 2024-04-18 20:29:53
[40888.220034] loop0: detected capacity change from 0 to 2097152
[40888.320215] nvmet: adding nsid 1 to subsystem blktests-subsystem-1
[40888.442496] nvmet_tcp: enabling port 0 (127.0.0.1:4420)
[40888.685458] nvmet: creating nvm controller 1 for subsystem
blktests-subsystem-1 for NQN
nqn.2014-08.org.nvmexpress:uuid:0f01fb42-9f7f-4856-b0b3-51e60b8de349.
[40888.703682] nvme nvme0: creating 48 I/O queues.
[40888.843827] nvme nvme0: mapped 48/0/0 default/read/poll queues.
[40888.999828] nvme nvme0: new ctrl: NQN "blktests-subsystem-1", addr
127.0.0.1:4420, hostnqn:
nqn.2014-08.org.nvmexpress:uuid:0f01fb42-9f7f-4856-b0b3-51e60b8de349
[40890.194368] XFS (nvme0n1): Mounting V5 Filesystem
064bbcf2-489b-4e7e-b264-a304e656990c
[40890.310950] XFS (nvme0n1): Ending clean mount
[40950.620522] nvme nvme0: I/O tag 49 (a031) type 4 opcode 0x1 (Write)
QID 25 timeout
[40950.628389] nvme nvme0: starting error recovery
[40950.633351] nvme nvme0: I/O tag 50 (8032) type 4 opcode 0x1 (Write)
QID 25 timeout
[40950.641915] nvme nvme0: I/O tag 51 (4033) type 4 opcode 0x1 (Write)
QID 25 timeout
[40950.649588] nvme nvme0: I/O tag 52 (1034) type 4 opcode 0x1 (Write)
QID 25 timeout
[40950.657313] block nvme0n1: no usable path - requeuing I/O
[40950.662841] block nvme0n1: no usable path - requeuing I/O
[40950.668484] nvme nvme0: I/O tag 53 (f035) type 4 opcode 0x1 (Write)
QID 25 timeout
[40950.676141] block nvme0n1: no usable path - requeuing I/O
[40950.681648] nvme nvme0: I/O tag 54 (a036) type 4 opcode 0x1 (Write)
QID 25 timeout
[40950.689323] block nvme0n1: no usable path - requeuing I/O
[40950.694913] block nvme0n1: no usable path - requeuing I/O
[40950.700372] nvme nvme0: I/O tag 55 (e037) type 4 opcode 0x1 (Write)
QID 25 timeout
[40950.707973] block nvme0n1: no usable path - requeuing I/O
[40950.713419] nvme nvme0: I/O tag 56 (6038) type 4 opcode 0x1 (Write)
QID 25 timeout
[40950.721007] block nvme0n1: no usable path - requeuing I/O
[40950.728055] nvme nvme0: I/O tag 57 (a039) type 4 opcode 0x1 (Write)
QID 25 timeout
[40950.735655] block nvme0n1: no usable path - requeuing I/O
[40950.741098] block nvme0n1: no usable path - requeuing I/O
[40950.746532] nvme nvme0: I/O tag 58 (c03a) type 4 opcode 0x1 (Write)
QID 25 timeout
[40950.754126] block nvme0n1: no usable path - requeuing I/O
[40950.760160] nvme nvme0: I/O tag 59 (303b) type 4 opcode 0x1 (Write)
QID 25 timeout
[40950.767822] nvme nvme0: I/O tag 60 (403c) type 4 opcode 0x1 (Write)
QID 25 timeout
[40950.775482] nvme nvme0: I/O tag 61 (703d) type 4 opcode 0x1 (Write)
QID 25 timeout
[40950.783139] nvme nvme0: I/O tag 62 (603e) type 4 opcode 0x1 (Write)
QID 25 timeout
[40950.790795] nvme nvme0: I/O tag 63 (303f) type 4 opcode 0x1 (Write)
QID 25 timeout
[40950.798465] nvme nvme0: I/O tag 64 (0040) type 4 opcode 0x1 (Write)
QID 25 timeout
[40950.806157] nvme nvme0: I/O tag 65 (1041) type 4 opcode 0x1 (Write)
QID 41 timeout
[40950.814373] nvme nvme0: I/O tag 66 (1042) type 4 opcode 0x1 (Write)
QID 41 timeout
[40950.822061] nvme nvme0: I/O tag 67 (1043) type 4 opcode 0x1 (Write)
QID 41 timeout
[40950.829735] nvme nvme0: I/O tag 68 (1044) type 4 opcode 0x1 (Write)
QID 41 timeout
[40950.837424] nvme nvme0: I/O tag 69 (1045) type 4 opcode 0x1 (Write)
QID 41 timeout
[40950.845108] nvme nvme0: I/O tag 70 (1046) type 4 opcode 0x1 (Write)
QID 41 timeout
[40950.852804] nvme nvme0: I/O tag 71 (1047) type 4 opcode 0x1 (Write)
QID 41 timeout
[40950.860677] nvme nvme0: I/O tag 80 (0050) type 4 opcode 0x0 (Flush)
QID 41 timeout
[40950.937368] nvme nvme0: Reconnecting in 10 seconds...
[40961.383043] nvmet: creating nvm controller 1 for subsystem
blktests-subsystem-1 for NQN
nqn.2014-08.org.nvmexpress:uuid:0f01fb42-9f7f-4856-b0b3-51e60b8de349.
[40961.394978] nvme nvme0: creating 48 I/O queues.
[40961.922944] nvme nvme0: mapped 48/0/0 default/read/poll queues.
[40962.017504] nvme nvme0: Successfully reconnected (1 attempt)
[41239.890559] XFS (nvme0n1): Unmounting Filesystem
064bbcf2-489b-4e7e-b264-a304e656990c
[41240.088647] nvme nvme0: Removing ctrl: NQN "blktests-subsystem-1"
[41329.981146] kworker/dying (319) used greatest stack depth: 18384 bytes left
[41733.311055] kmemleak: 46 new suspected memory leaks (see
/sys/kernel/debug/kmemleak)
>
> > # dmesg | grep kmemleak
> > [ 2580.572467] kmemleak: 92 new suspected memory leaks (see
> > /sys/kernel/debug/kmemleak)
> >
> > # cat kmemleak.log
> > unreferenced object 0xffff8885a1abe740 (size 32):
> > comm "kworker/40:1H", pid 799, jiffies 4296062986
> > hex dump (first 32 bytes):
> > c2 4a 4a 04 00 ea ff ff 00 00 00 00 00 10 00 00 .JJ.............
> > 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 ................
> > backtrace (crc 6328eade):
> > [<ffffffffa7f2657c>] __kmalloc+0x37c/0x480
> > [<ffffffffa86a9b1f>] sgl_alloc_order+0x7f/0x360
> > [<ffffffffc261f6c5>] lo_read_simple+0x1d5/0x5b0 [loop]
> > [<ffffffffc26287ef>] 0xffffffffc26287ef
> > [<ffffffffc262a2c4>] 0xffffffffc262a2c4
> > [<ffffffffc262a881>] 0xffffffffc262a881
> > [<ffffffffa76adf3c>] process_one_work+0x89c/0x19f0
> > [<ffffffffa76b0813>] worker_thread+0x583/0xd20
> > [<ffffffffa76ce2a3>] kthread+0x2f3/0x3e0
> > [<ffffffffa74a804d>] ret_from_fork+0x2d/0x70
> > [<ffffffffa7406e4a>] ret_from_fork_asm+0x1a/0x30
> > unreferenced object 0xffff88a8b03647c0 (size 16):
> > comm "kworker/40:1H", pid 799, jiffies 4296062986
> > hex dump (first 16 bytes):
> > c0 4a 4a 04 00 ea ff ff 00 10 00 00 00 00 00 00 .JJ.............
> > backtrace (crc 860ce62b):
> > [<ffffffffa7f2657c>] __kmalloc+0x37c/0x480
> > [<ffffffffc261f805>] lo_read_simple+0x315/0x5b0 [loop]
> > [<ffffffffc26287ef>] 0xffffffffc26287ef
> > [<ffffffffc262a2c4>] 0xffffffffc262a2c4
> > [<ffffffffc262a881>] 0xffffffffc262a881
> > [<ffffffffa76adf3c>] process_one_work+0x89c/0x19f0
> > [<ffffffffa76b0813>] worker_thread+0x583/0xd20
> > [<ffffffffa76ce2a3>] kthread+0x2f3/0x3e0
> > [<ffffffffa74a804d>] ret_from_fork+0x2d/0x70
> > [<ffffffffa7406e4a>] ret_from_fork_asm+0x1a/0x30
> >
> > (gdb) l *(lo_read_simple+0x1d5)
> > 0x66c5 is in lo_read_simple (drivers/block/loop.c:284).
> > 279 struct bio_vec bvec;
> > 280 struct req_iterator iter;
> > 281 struct iov_iter i;
> > 282 ssize_t len;
> > 283
> > 284 rq_for_each_segment(bvec, rq, iter) {
> > 285 iov_iter_bvec(&i, ITER_DEST, &bvec, 1, bvec.bv_len);
> > 286 len = vfs_iter_read(lo->lo_backing_file, &i, &pos, 0);
> > 287 if (len < 0)
> > 288 return len;
> > (gdb) l *(lo_read_simple+0x315)
> > 0x6805 is in lo_read_simple (./include/linux/bio.h:120).
> > 115 iter->bi_sector += bytes >> 9;
> > 116
> > 117 if (bio_no_advance_iter(bio))
> > 118 iter->bi_size -= bytes;
> > 119 else
> > 120 bvec_iter_advance_single(bio->bi_io_vec, iter, bytes);
> > 121 }
> > 122
> > 123 void __bio_advance(struct bio *, unsigned bytes);
> > 124
> >
> >
>
> -ck
>
>
--
Best Regards,
Yi Zhang
More information about the Linux-nvme
mailing list