Kernel failure after update to 2.6.27

zak_wang · #1 23rd November 2008, 06:54 AM

I updated my Fedora 9 kernel to 2.6.27 the other day and received message of "kernel failure". However, I didn't see any defects brought along.

I sent the report to kerneloops.org and switched back to kernel 2.6.26. Down here, I tried liveCD of Fedora 10 Preview on virtualbox----kernel failure came up again. So I downloaded the full installation iso of F10PR and did the install. Same thing happened.

So I was assured, it's the kernel. Now I began to worry whether I could update to Fedora 10 on Tuesday because it provides no 2.6.26.

Below is the trace report:

Kernel failure message 1:
------------[ cut here ]------------
WARNING: at arch/x86/kernel/cpu/mtrr/main.c:1500 mtrr_trim_uncached_memo
WARNING: strange, CPU MTRRs all blank?
Modules linked in:
Pid: 0, comm: swapper Not tainted 2.6.27.4-68.fc10.i686 #1
[<c042ba36>] warn_slowpath+0x4b/0x6c
[<c0520cbd>] ? vsnprintf+0x340/0x54e
[<c06c5ffc>] ? _spin_unlock_irqrestore+0x1b/0x38
[<c06c5ffc>] ? _spin_unlock_irqrestore+0x1b/0x38
[<c06c6003>] ? _spin_unlock_irqrestore+0x22/0x38
[<c042c1a1>] ? release_console_sem+0x1bf/0x1c7
[<c040d9da>] ? mtrr_wrmsr+0x12/0x35
[<c084043c>] ? get_mtrr_state+0x299/0x2bb
[<c0840443>] ? get_mtrr_state+0x2a0/0x2bb
[<c08400c6>] mtrr_trim_uncached_memory+0x30e/0x315
[<c043b529>] ? parse_args+0x143/0x1d2
[<c06c41a7>] ? printk+0xf/0x18
[<c083ccd0>] ? e820_end_pfn+0xa2/0xac
[<c083b667>] setup_arch+0x47b/0x938
[<c042c61c>] ? vprintk+0x2ce/0x2f3
[<c083d80f>] ? __reserve_early+0xe4/0xf8
[<c06c41a7>] ? printk+0xf/0x18
[<c0835605>] start_kernel+0x73/0x2f0
[<c0835080>] __init_begin+0x80/0x88
=======================
---[ end trace 4eaa2a86a8e2da22 ]---

marko · #2 23rd November 2008, 07:29 AM

I think this only happens in virtualization, that is in VBox, KVM and VmWare the "MTRR" is blank but that won't be the case with real hardware. So you'll probably be fine on your real PC and this is a issue with Vbox.

What I'm not sure of is whether your first "kernel failure" was related or
something else, was that the same mtrr error?

Here's a bit of some kernel developers talking about this problem:

REF:

http://groups.google.com/group/linux...ce00c?lnk=raot

zak_wang · #3 23rd November 2008, 02:55 PM

Thanks. With that said, I dived into the real machine with kernel 2.6.27.5-37.fc9.i686 and received this kernel failure:

Kernel failure message 1:
------------[ cut here ]------------
WARNING: at fs/sysfs/dir.c:463 sysfs_add_one+0x2a/0x36()
sysfs: duplicate filename '1' can not be created
Modules linked in:
Pid: 1, comm: swapper Not tainted 2.6.27.5-37.fc9.i686 #1
[<c042a052>] warn_slowpath+0x4b/0x6c
[<c0402707>] ? __switch_to+0xd3/0x153
[<c0500684>] ? idr_get_empty_slot+0x12a/0x1d9
[<c0500803>] ? ida_get_new_above+0xd0/0x171
[<c04c9ef4>] ? sysfs_ilookup_test+0x0/0x11
[<c049f0ee>] ? ifind+0x55/0x60
[<c04ca0e7>] ? sysfs_find_dirent+0x16/0x27
[<c04ca239>] sysfs_add_one+0x2a/0x36
[<c04ca6d6>] create_dir+0x43/0x68
[<c04ca728>] sysfs_create_dir+0x2d/0x41
[<c0501185>] kobject_add_internal+0xb2/0x153
[<c05012d1>] kobject_add_varg+0x35/0x41
[<c05012fd>] kobject_init_and_add+0x20/0x22
[<c050f172>] pci_create_slot+0x93/0xe1
[<c054a3a8>] register_slot+0x14f/0x1e7
[<c043e958>] ? up+0x2b/0x2f
[<c053b99e>] acpi_ns_walk_namespace+0x9a/0x110
[<c053a271>] acpi_walk_namespace+0x46/0x66
[<c054a259>] ? register_slot+0x0/0x1e7
[<c054a228>] walk_p2p_bridge+0xb7/0xe8
[<c054a259>] ? register_slot+0x0/0x1e7
[<c054a259>] ? register_slot+0x0/0x1e7
[<c054a259>] ? register_slot+0x0/0x1e7
[<c053b99e>] acpi_ns_walk_namespace+0x9a/0x110
[<c053a271>] acpi_walk_namespace+0x46/0x66
[<c054a171>] ? walk_p2p_bridge+0x0/0xe8
[<c054a124>] acpi_pci_slot_add+0x10d/0x15a
[<c054a171>] ? walk_p2p_bridge+0x0/0xe8
[<c054a259>] ? register_slot+0x0/0x1e7
[<c0548901>] acpi_pci_register_driver+0x39/0x50
[<c079f5f6>] acpi_pci_slot_init+0x17/0x1b
[<c0401125>] _stext+0x3d/0x115
[<c079f5df>] ? acpi_pci_slot_init+0x0/0x1b
[<c0463028>] ? register_irq_proc+0x92/0xae
[<c0463081>] ? init_irq_proc+0x3d/0x4a
[<c077e362>] kernel_init+0x177/0x1c5
[<c077e1eb>] ? kernel_init+0x0/0x1c5
[<c040494b>] kernel_thread_helper+0x7/0x10
=======================
---[ end trace a7919e7f17c0a725 ]---

I'm not familiar with this kernel thing but I saw a token '-EEXIST' appered in the message printed out while booting kernel. Any clues?
Thanks again.

aleph · #4 23rd November 2008, 03:35 PM

I'm using kernel 2.6.27.7-50.fc9.i686 and there's no problem with that. Can you try the latest Koji build and see if the problem goes away?

MALDATA · #5 23rd November 2008, 04:14 PM

I had a similar problem... the rawhide kernels crashed my machine. I filed a bug report here:

https://bugzilla.redhat.com/show_bug.cgi?id=465684

And stumbled across a workaround. If I add

Code:

console=ttyS0,9600

to my kernel line, it boots. You don't actually need to hook up a serial cable, just adding the line above makes it work for me.

Of course, we're probably having different problems, so I can't guarantee it'll work, but it'll only take you 2 minutes to try it. Hopefully it'll help.

zak_wang · #6 24th November 2008, 01:19 AM

Thanks maldata. I went to your bugreport on bugzilla. Our situations are actually quite different. Except for the notification to the right-bottom corner of my desktop, I'm not really suffering from any defects of the system. I'm trying to update my kernel.

Besides, I printed the system messages sincing booting using

Code:

dmsg|less

and cut part of it:

Code:

ACPI: AC Adapter [AC] (on-line)
ACPI: Battery Slot [BAT0] (battery present)
input: Power Button (FF) as /devices/LNXSYSTM:00/LNXPWRBN:00/input/input0
ACPI: Power Button (FF) [PWRF]
input: Lid Switch as /devices/LNXSYSTM:00/device:00/PNP0C0D:00/input/input1
ACPI: Lid Switch [LID]
input: Sleep Button (CM) as /devices/LNXSYSTM:00/device:00/PNP0C0E:00/input/input2
ACPI: Sleep Button (CM) [SLPB]
------------[ cut here ]------------
WARNING: at fs/sysfs/dir.c:463 sysfs_add_one+0x2a/0x36()
sysfs: duplicate filename '1' can not be created
Modules linked in:
Pid: 1, comm: swapper Not tainted 2.6.27.5-37.fc9.i686 #1
 [<c042a052>] warn_slowpath+0x4b/0x6c
 [<c0402707>] ? __switch_to+0xd3/0x153
 [<c0500684>] ? idr_get_empty_slot+0x12a/0x1d9
 [<c0500803>] ? ida_get_new_above+0xd0/0x171
 [<c04c9ef4>] ? sysfs_ilookup_test+0x0/0x11
 [<c049f0ee>] ? ifind+0x55/0x60
 [<c04ca0e7>] ? sysfs_find_dirent+0x16/0x27
 [<c04ca239>] sysfs_add_one+0x2a/0x36
 [<c04ca6d6>] create_dir+0x43/0x68
 [<c04ca728>] sysfs_create_dir+0x2d/0x41
 [<c0501185>] kobject_add_internal+0xb2/0x153
 [<c05012d1>] kobject_add_varg+0x35/0x41
 [<c05012fd>] kobject_init_and_add+0x20/0x22
 [<c050f172>] pci_create_slot+0x93/0xe1
 [<c054a3a8>] register_slot+0x14f/0x1e7
 [<c043e958>] ? up+0x2b/0x2f
 [<c053b99e>] acpi_ns_walk_namespace+0x9a/0x110
 [<c053a271>] acpi_walk_namespace+0x46/0x66
 [<c054a259>] ? register_slot+0x0/0x1e7
 [<c054a228>] walk_p2p_bridge+0xb7/0xe8
 [<c054a259>] ? register_slot+0x0/0x1e7
 [<c054a259>] ? register_slot+0x0/0x1e7
 [<c054a259>] ? register_slot+0x0/0x1e7
 [<c053b99e>] acpi_ns_walk_namespace+0x9a/0x110
 [<c053a271>] acpi_walk_namespace+0x46/0x66
 [<c054a171>] ? walk_p2p_bridge+0x0/0xe8
 [<c054a124>] acpi_pci_slot_add+0x10d/0x15a
 [<c054a171>] ? walk_p2p_bridge+0x0/0xe8
 [<c054a259>] ? register_slot+0x0/0x1e7
 [<c054a259>] ? register_slot+0x0/0x1e7
 [<c054a259>] ? register_slot+0x0/0x1e7
 [<c053b99e>] acpi_ns_walk_namespace+0x9a/0x110
 [<c053a271>] acpi_walk_namespace+0x46/0x66
 [<c054a171>] ? walk_p2p_bridge+0x0/0xe8
 [<c054a124>] acpi_pci_slot_add+0x10d/0x15a
 [<c054a171>] ? walk_p2p_bridge+0x0/0xe8
 [<c054a259>] ? register_slot+0x0/0x1e7
 [<c0548901>] acpi_pci_register_driver+0x39/0x50
 [<c079f5f6>] acpi_pci_slot_init+0x17/0x1b
 [<c0401125>] _stext+0x3d/0x115
 [<c079f5df>] ? acpi_pci_slot_init+0x0/0x1b
 [<c0463028>] ? register_irq_proc+0x92/0xae
 [<c0463081>] ? init_irq_proc+0x3d/0x4a
 [<c077e362>] kernel_init+0x177/0x1c5
 [<c077e1eb>] ? kernel_init+0x0/0x1c5
 [<c040494b>] kernel_thread_helper+0x7/0x10
 =======================
---[ end trace a7919e7f17c0a725 ]---
kobject_add_internal failed for 1 with -EEXIST, don't try to register things with the same name in the same directory.
Pid: 1, comm: swapper Tainted: G        W 2.6.27.5-37.fc9.i686 #1
 [<c0501214>] kobject_add_internal+0x141/0x153
 [<c05012d1>] kobject_add_varg+0x35/0x41
 [<c05012fd>] kobject_init_and_add+0x20/0x22
 [<c050f172>] pci_create_slot+0x93/0xe1
 [<c054a3a8>] register_slot+0x14f/0x1e7
 [<c043e958>] ? up+0x2b/0x2f
 [<c053b99e>] acpi_ns_walk_namespace+0x9a/0x110
 [<c053a271>] acpi_walk_namespace+0x46/0x66
 [<c054a259>] ? register_slot+0x0/0x1e7
 [<c054a228>] walk_p2p_bridge+0xb7/0xe8
 [<c054a259>] ? register_slot+0x0/0x1e7
 [<c054a259>] ? register_slot+0x0/0x1e7
 [<c054a259>] ? register_slot+0x0/0x1e7
 [<c053b99e>] acpi_ns_walk_namespace+0x9a/0x110
 [<c053a271>] acpi_walk_namespace+0x46/0x66
 [<c054a171>] ? walk_p2p_bridge+0x0/0xe8
 [<c054a124>] acpi_pci_slot_add+0x10d/0x15a
 [<c054a171>] ? walk_p2p_bridge+0x0/0xe8
 [<c054a259>] ? register_slot+0x0/0x1e7
 [<c0548901>] acpi_pci_register_driver+0x39/0x50
 [<c079f5f6>] acpi_pci_slot_init+0x17/0x1b
 [<c0401125>] _stext+0x3d/0x115
 [<c079f5df>] ? acpi_pci_slot_init+0x0/0x1b
 [<c0463028>] ? register_irq_proc+0x92/0xae
 [<c0463081>] ? init_irq_proc+0x3d/0x4a
 [<c077e362>] kernel_init+0x177/0x1c5
 [<c077e1eb>] ? kernel_init+0x0/0x1c5
 [<c040494b>] kernel_thread_helper+0x7/0x10
 =======================
Unable to register kobject 1
pci_slot: pci_create_slot returned -17

I'll come back to report results of my update.

zak_wang · #7 24th November 2008, 08:05 AM

I thought this new kernel which had its last number changed did it because the problem was fixed. But it wasn't. After the update, same thing happened. Kernel version is now 2.6.27.5-41.fc9.i686.

kobject_add_internal failed for 1 with -EEXIST, don't try to register things with the same name in the same directory.
Pid: 1, comm: swapper Tainted: G W 2.6.27.5-37.fc9.i686 #1