can't start qemu-kvm on 2.6.34-rc3

can't start qemu-kvm on 2.6.34-rc3

[Tomasz Chmielewski]

With qemu-kvm 0.12.3 used on 2.6.34-rc3, this command:

qemu-kvm -m 1500 -drive file=/srv/kvm/images/im1.qcow2,if=virtio,cache=none,index=0,boot=on -drive file=/srv/kvm/images/im1-backup.qcow2,if=virtio,cache=none,index=1 -net nic,vlan=0,model=virtio,macaddr=F2:4A:51:41:B1:AA -net tap,vlan=0,script=/etc/qemu-ifup -localtime -nographic

Renders the below - is it a known issue, or something particular with my configuration?


[  282.364859] BUG: unable to handle kernel paging request at 0000000200000001
[  282.364863] IP: [<ffffffff8111c805>] __kmalloc_node+0x125/0x200
[  282.364869] PGD 17d967067 PUD 0 
[  282.364871] Oops: 0000 [#1] SMP 
[  282.364873] last sysfs file: /sys/devices/system/cpu/cpu7/cache/index2/shared_cpu_map
[  282.364875] CPU 3 
[  282.364876] Modules linked in: bridge stp radeon ttm drm_kms_helper drm i2c_algo_bit tun af_packet xt_tcpudp iptable_filter ip_tables x_tables ipv6 coretemp binfmt_misc loop dm_mod cpufreq_conservative cpufreq_powersave acpi_cpufreq kvm_intel kvm snd_hda_codec_atihdmi snd_hda_codec_realtek snd_hda_intel snd_hda_codec snd_hwdep snd_seq_dummy snd_seq_oss snd_seq_midi_event snd_seq snd_seq_device snd_pcm_oss snd_pcm snd_timer snd_mixer_oss joydev iTCO_wdt snd soundcore snd_page_alloc wmi processor evdev i2c_i801 iTCO_vendor_support i2c_core sr_mod e1000e sg pcspkr thermal button serio_raw ata_piix ahci libata sd_mod scsi_mod crc_t10dif raid1 ext4 jbd2 crc16 uhci_hcd ohci_hcd ehci_hcd usbhid hid usbcore [last unloaded: scsi_wait_scan]
[  282.364908] 
[  282.364909] Pid: 14874, comm: qemu-kvm Not tainted 2.6.34-rc3 #1 DX58SO/        
[  282.364911] RIP: 0010:[<ffffffff8111c805>]  [<ffffffff8111c805>] __kmalloc_node+0x125/0x200
[  282.364914] RSP: 0018:ffff88017e983ae8  EFLAGS: 00010046
[  282.364916] RAX: ffff880001a72568 RBX: 0000000200000001 RCX: ffffffff81106153
[  282.364917] RDX: 0000000000000000 RSI: 00000000000080d0 RDI: 0000000000000003
[  282.364918] RBP: ffff88017e983b38 R08: ffffffffa02fff5c R09: 00000000000000d2
[  282.364920] R10: 0000000000000001 R11: 0000000000000001 R12: ffffffff8160db68
[  282.364921] R13: 00000000000080d0 R14: 00000000000080d0 R15: 0000000000000246
[  282.364923] FS:  00007f838f566710(0000) GS:ffff880001a60000(0000) knlGS:0000000000000000
[  282.364925] CS:  0010 DS: 002b ES: 002b CR0: 000000008005003b
[  282.364926] CR2: 0000000200000001 CR3: 000000017db7e000 CR4: 00000000000026e0
[  282.364928] DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
[  282.364929] DR3: 0000000000000000 DR6: 00000000ffff0ff0 DR7: 0000000000000400
[  282.364931] Process qemu-kvm (pid: 14874, threadinfo ffff88017e982000, task ffff88017e478000)
[  282.364932] Stack:
[  282.364933]  0000000000000000 ffffffff81106153 ffffffffffffffff 0000000000000008
[  282.364935] <0> ffff88023d94d440 ffff88023d94d440 00000000ffffffff ffffffffa02fff5c
[  282.364937] <0> 8000000000000163 00000000000000d2 ffff88017e983b98 ffffffff81106153
[  282.364940] Call Trace:
[  282.364944]  [<ffffffff81106153>] ? __vmalloc_area_node+0x63/0x190
[  282.364955]  [<ffffffffa02fff5c>] ? __kvm_set_memory_region+0x61c/0x7a0 [kvm]
[  282.364957]  [<ffffffff81106153>] __vmalloc_area_node+0x63/0x190
[  282.364963]  [<ffffffffa02fff5c>] ? __kvm_set_memory_region+0x61c/0x7a0 [kvm]
[  282.364966]  [<ffffffff811060e2>] __vmalloc_node+0xa2/0xb0
[  282.364971]  [<ffffffffa02fff5c>] ? __kvm_set_memory_region+0x61c/0x7a0 [kvm]
[  282.364974]  [<ffffffff8110643c>] vmalloc+0x2c/0x30
[  282.364979]  [<ffffffffa02fff5c>] __kvm_set_memory_region+0x61c/0x7a0 [kvm]
[  282.364984]  [<ffffffffa02fc1c8>] ? kvm_io_bus_write+0x68/0xa0 [kvm]
[  282.364991]  [<ffffffffa0300123>] kvm_set_memory_region+0x43/0x70 [kvm]
[  282.364997]  [<ffffffffa030016d>] kvm_vm_ioctl_set_memory_region+0x1d/0x30 [kvm]
[  282.365003]  [<ffffffffa03003f0>] kvm_vm_ioctl+0x270/0x410 [kvm]
[  282.365009]  [<ffffffffa03014ee>] ? kvm_dev_ioctl+0xbe/0x440 [kvm]
[  282.365011]  [<ffffffff8113572d>] vfs_ioctl+0x3d/0xd0
[  282.365013]  [<ffffffff81135cba>] do_vfs_ioctl+0x8a/0x5a0
[  282.365016]  [<ffffffff811c2a55>] ? tomoyo_path_perm+0x45/0x110
[  282.365018]  [<ffffffff81136251>] sys_ioctl+0x81/0xa0
[  282.365021]  [<ffffffff8100a002>] system_call_fastpath+0x16/0x1b
[  282.365023] Code: e8 4c 8b 75 f0 4c 8b 7d f8 c9 c3 0f 1f 00 49 63 54 24 14 31 f6 48 89 df e8 49 89 0d 00 eb bc 0f 1f 80 00 00 00 00 49 63 54 24 18 <48> 8b 14 13 48 89 10 eb 92 66 90 48 89 4d b8 e8 77 ce 29 00 48 
[  282.365041] RIP  [<ffffffff8111c805>] __kmalloc_node+0x125/0x200
[  282.365043]  RSP <ffff88017e983ae8>
[  282.365044] CR2: 0000000200000001
[  282.365046] ---[ end trace 7fb0c79c903996ce ]---



-- 
Tomasz Chmielewski

Re: can't start qemu-kvm on 2.6.34-rc3

[Avi Kivity]

On 03/31/2010 01:39 AM, Tomasz Chmielewski wrote:
> With qemu-kvm 0.12.3 used on 2.6.34-rc3, this command:
>
> qemu-kvm -m 1500 -drive file=/srv/kvm/images/im1.qcow2,if=virtio,cache=none,index=0,boot=on -drive file=/srv/kvm/images/im1-backup.qcow2,if=virtio,cache=none,index=1 -net nic,vlan=0,model=virtio,macaddr=F2:4A:51:41:B1:AA -net tap,vlan=0,script=/etc/qemu-ifup -localtime -nographic
>
> Renders the below - is it a known issue, or something particular with my configuration?
>
>    

Looks unrelated to kvm, rather a problem in the vmalloc subsystem.  Can 
you bisect it?  If you have an AMD machine, you can bisect inside a 
guest using -enable-nesting, which should reduce the pain involved.


-- 
Do not meddle in the internals of kernels, for they are subtle and quick to panic.

Re: can't start qemu-kvm on 2.6.34-rc3

[Tomasz Chmielewski]

> With qemu-kvm 0.12.3 used on 2.6.34-rc3, this command:
>
> qemu-kvm -m 1500 -drive file=/srv/kvm/images/im1.qcow2,if=virtio,cache=none,index=0,boot=on
> -drive file=/srv/kvm/images/im1-backup.qcow2,if=virtio,cache=none,index=1 -net
> nic,vlan=0,model=virtio,macaddr=F2:4A:51:41:B1:AA -net tap,vlan=0,script=/etc/qemu-ifup
> -localtime -nographic
>
> Renders the below - is it a known issue, or something particular with my configuration?
>
> [  282.364859] BUG: unable to handle kernel paging request at 0000000200000001
> [  282.364863] IP: [<ffffffff8111c805>] __kmalloc_node+0x125/0x200
> [  282.364869] PGD 17d967067 PUD 0
> [  282.364871] Oops: 0000 [#1] SMP
> [  282.364873] last sysfs file: /sys/devices/system/cpu/cpu7/cache/index2/shared_cpu_map

(...)

If it's interesting to anyone, kvm starts fine here if I start it 
_before_ starting X.

If I first start X, then kvm - machines starts to Oops.


-- 
Tomasz Chmielewski
http://wpkg.org