1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
|
user-level: 0.881
architecture: 0.856
permissions: 0.847
mistranslation: 0.835
KVM: 0.824
device: 0.818
arm: 0.815
register: 0.815
graphic: 0.812
virtual: 0.806
TCG: 0.806
peripherals: 0.805
assembly: 0.805
debug: 0.805
VMM: 0.803
vnc: 0.801
x86: 0.798
files: 0.796
socket: 0.795
performance: 0.789
PID: 0.789
i386: 0.774
kernel: 0.771
network: 0.770
ppc: 0.766
semantic: 0.751
risc-v: 0.750
hypervisor: 0.745
boot: 0.738
Qemu 2.2.0 savevm crash.
qemu 2.1.2 is good.
(gdb) bt
#0 0x00007ffff4aae445 in raise () from /lib/x86_64-linux-gnu/libc.so.6
#1 0x00007ffff4ab1bab in abort () from /lib/x86_64-linux-gnu/libc.so.6
#2 0x0000555555997951 in qcow2_cache_find_entry_to_replace (c=0x555556389780) at block/qcow2-cache.c:262
#3 0x0000555555997a1a in qcow2_cache_do_get (bs=0x5555563836f0, c=0x555556389780, offset=1094713344, table=0x7fffffffc528,
read_from_disk=true) at block/qcow2-cache.c:285
#4 0x0000555555997c45 in qcow2_cache_get (bs=0x5555563836f0, c=0x555556389780, offset=1094713344, table=0x7fffffffc528)
at block/qcow2-cache.c:331
#5 0x0000555555991ca9 in l2_allocate (bs=0x5555563836f0, l1_index=1, table=0x7fffffffc5a0) at block/qcow2-cluster.c:247
#6 0x000055555599290c in get_cluster_table (bs=0x5555563836f0, offset=549755813888, new_l2_table=0x7fffffffc610,
new_l2_index=0x7fffffffc62c) at block/qcow2-cluster.c:620
#7 0x0000555555994213 in discard_single_l2 (bs=0x5555563836f0, offset=549755813888, nb_clusters=156, type=QCOW2_DISCARD_NEVER,
full_discard=false) at block/qcow2-cluster.c:1425
#8 0x0000555555994491 in qcow2_discard_clusters (bs=0x5555563836f0, offset=549755813888, nb_sectors=638976, type=QCOW2_DISCARD_NEVER,
full_discard=false) at block/qcow2-cluster.c:1516
#9 0x00005555559965c8 in qcow2_snapshot_create (bs=0x5555563836f0, sn_info=0x7fffffffc830) at block/qcow2-snapshot.c:441
#10 0x00005555559ad1ad in bdrv_snapshot_create (bs=0x5555563836f0, sn_info=0x7fffffffc830) at block/snapshot.c:167
#11 0x000055555565e90f in do_savevm (mon=0x555556992d20, qdict=0x5555599d5c00) at /vms/qemu/qemu-2.2.0/savevm.c:1126
(gdb) show args
Argument list to give program being debugged when it is started is "-name u1404-01 -S -machine pc,accel=kvm,usb=off -m 1024 -smp 2,sockets=2,cores=1,threads=1 -no-user-config -nodefaults -monitor stdio -rtc base=utc -no-shutdown -device piix3-usb-uhci,id=usb,bus=pci.0,addr=0x1.0x2 -device virtio-serial-pci,id=virtio-serial0,bus=pci.0,addr=0x5 -drive file=/vms/images/u1404-01.img,if=none,id=drive-virtio-disk0,format=qcow2,cache=directsync -device virtio-blk-pci,scsi=off,bus=pci.0,addr=0x4,drive=drive-virtio-disk0,id=virtio-disk0 -drive if=none,id=drive-ide0-1-0,readonly=on,format=raw -device ide-cd,bus=ide.1,unit=0,drive=drive-ide0-1-0,id=ide0-1-0 -chardev pty,id=charserial0 -device isa-serial,chardev=charserial0,id=serial0 -device usb-tablet,id=input0 -vnc 0.0.0.0:0 -device cirrus-vga,id=video0,bus=pci.0,addr=0x2 -device virtio-balloon-pci,id=balloon0,bus=pci.0,addr=0x6".
Maybe bdrv_snapshot_create() should take s->lock but it's not clear yet what causes all qcow2 cache entries to be referenced.
How do you reproduce this crash? Please give exact steps including what commands to run inside the guest and what QEMU monitor commands to run.
Is the crash deterministic (does it happen every time or with a random chance)?
The bug can be reproduced every time.
1. I install a Ubuntu 14.04 guest.
2. The monitor command is (qemu) snapshot_blkdev_internal drive-virtio-disk0 s1
3. Or (qemu) savevm s1
I think I get the reason.
From Qemu 2.2.0, in qcow2_open: l2_cache_size = MAX(DEFAULT_L2_CACHE_BYTE_SIZE/s->cluster_size, MIN_L2_CACHE_SIZE);
Here DEFAULT_L2_CACHE_BYTE_SIZE=1M, MIN_L2_CACHE_SIZE=1.
If the cluster_size < 1M, the l2_cache_size will be greater than 1, it is ok.
If the cluster_size>=1M, the l2_cache_size will be 1. After create snapshot, the first qcow2_co_writev will abort at qcow2_cache_find_entry_to_replace, because no free room in l2_table_cache. If change the MIN_L2_CACHE_SIZE to 16, it is ok.
Triaging old bug tickets... can you still reproduce this issue with the latest version of QEMU? Or could we close this ticket nowadays?
|