2.6 git kernel reporting bug in knodemgrd_0 during boot

2.6 git kernel reporting bug in knodemgrd_0 during boot

[Andrew Paprocki]

I just upgraded my dev box to the latest 2.6 source via git and this
is now printing out in dmesg upon every boot. Hardware is a standard
VIA EPIA-MII motherboard w/ nothing extra added. I can supply
additional info if anyone requests it. -Andrew

Probing IDE interface ide0...
hda: HITACHI_DK23DA-40, ATA DISK drive
ieee1394: Host added: ID:BUS[0-00:1023]  GUID[0040635000015f17]
------------[ cut here ]------------
Kernel BUG at [verbose debug info unavailable]
invalid opcode: 0000 [#1]
Modules linked in:
CPU:    0
EIP:    0060:[<b014932e>]    Not tainted VLI
EFLAGS: 00010002   (2.6.19-rc3-g2da6dc28 #1)
EIP is at cache_alloc_refill+0x18e/0x400
eax: 00000000   ebx: 0000003c   ecx: 00000000   edx: 00000000
esi: bdfed770   edi: bdfed760   ebp: bdfea000   esp: bd2bbe00
ds: 007b   es: 007b   ss: 0068
Process knodemgrd_0 (pid: 174, ti=bd2ba000 task=bdfa8ab0 task.ti=bd2ba000)
Stack: 000200d2 000000d2 bdfef0c0 bdfee1e0 0000003c 00000000 00000001 b03a61ec
       000000d2 b03a61ec bdfa8ab0 00000246 00001fff 000000d2 00000163 b0149192
       be800000 b0143085 bd2bbe7c 00000002 ffffffff 00000001 000000d2 ffffffff
Call Trace:
 [<b0149192>] kmem_cache_alloc+0x22/0x30
 [<b0143085>] __get_vm_area_node+0x95/0x160
 [<b0143187>] get_vm_area_node+0x37/0x40
 [<b014370c>] __vmalloc_node+0x3c/0x60
 [<b014375f>] __vmalloc+0xf/0x20
 [<b0269ced>] csr1212_attach_keyval_to
_directory+0x1d/0x60
 [<b026a52c>] csr1212_parse_keyval+0x14c/0x200
 [<b026a9be>] _csr1212_read_keyval+0x3de/0x420
 [<b0268185>] nodemgr_probe_ne+0x205/0x390
 [<b0179561>] sysfs_add_file+0x61/0x70
 [<b0268c3c>] nodemgr_host_thread+0x82c/0x980
 [<b0268410>] nodemgr_host_thread+0x0/0x980
 [<b0125a4f>] kthread+0xaf/0xe0
 [<b01259a0>] kthread+0x0/0xe0
 [<b0103d47>] kernel_thread_helper+0x7/0x10
 =======================
Code: 04 89 37 83 7c 24 10 00 0f 8f 5a ff ff ff 8b 47 18 2b 45 00 89
47 18 83 7d 00 00 0f 85 61 02 00 00 f7 44 24 04 0e 80 f8 ff 74 02 <0f>
0b f7 44 24 04 00 20 00 00 0f 85 2d 02 00 00 8b 5c 24 04 81
EIP: [<b014932e>] cache_alloc_refill+0x18e/0x400 SS:ESP 0068:bd2bbe00
 ide0 at 0x1f0-0x1f7,0x3f6 on irq 14

Re: 2.6 git kernel reporting bug in knodemgrd_0 during boot

[Adrian Bunk]

On Sun, Oct 29, 2006 at 04:41:54PM -0500, Andrew Paprocki wrote:

> I just upgraded my dev box to the latest 2.6 source via git and this
> is now printing out in dmesg upon every boot. Hardware is a standard
> VIA EPIA-MII motherboard w/ nothing extra added. I can supply
> additional info if anyone requests it. -Andrew

The ieee1394 maintainers (Cc'ed) might be the best contact.

> Probing IDE interface ide0...
> hda: HITACHI_DK23DA-40, ATA DISK drive
> ieee1394: Host added: ID:BUS[0-00:1023]  GUID[0040635000015f17]
> ------------[ cut here ]------------
> Kernel BUG at [verbose debug info unavailable]

The bug message with CONFIG_DEBUG_BUGVERBOSE=y set might be interesting.

> invalid opcode: 0000 [#1]
> Modules linked in:
> CPU:    0
> EIP:    0060:[<b014932e>]    Not tainted VLI
> EFLAGS: 00010002   (2.6.19-rc3-g2da6dc28 #1)
> EIP is at cache_alloc_refill+0x18e/0x400
> eax: 00000000   ebx: 0000003c   ecx: 00000000   edx: 00000000
> esi: bdfed770   edi: bdfed760   ebp: bdfea000   esp: bd2bbe00
> ds: 007b   es: 007b   ss: 0068
> Process knodemgrd_0 (pid: 174, ti=bd2ba000 task=bdfa8ab0 task.ti=bd2ba000)
> Stack: 000200d2 000000d2 bdfef0c0 bdfee1e0 0000003c 00000000 00000001 
> b03a61ec
>       000000d2 b03a61ec bdfa8ab0 00000246 00001fff 000000d2 00000163 
>       b0149192
>       be800000 b0143085 bd2bbe7c 00000002 ffffffff 00000001 000000d2 
>       ffffffff
> Call Trace:
> [<b0149192>] kmem_cache_alloc+0x22/0x30
> [<b0143085>] __get_vm_area_node+0x95/0x160
> [<b0143187>] get_vm_area_node+0x37/0x40
> [<b014370c>] __vmalloc_node+0x3c/0x60
> [<b014375f>] __vmalloc+0xf/0x20
> [<b0269ced>] csr1212_attach_keyval_to
> _directory+0x1d/0x60
> [<b026a52c>] csr1212_parse_keyval+0x14c/0x200
> [<b026a9be>] _csr1212_read_keyval+0x3de/0x420
> [<b0268185>] nodemgr_probe_ne+0x205/0x390
> [<b0179561>] sysfs_add_file+0x61/0x70
> [<b0268c3c>] nodemgr_host_thread+0x82c/0x980
> [<b0268410>] nodemgr_host_thread+0x0/0x980
> [<b0125a4f>] kthread+0xaf/0xe0
> [<b01259a0>] kthread+0x0/0xe0
> [<b0103d47>] kernel_thread_helper+0x7/0x10
> =======================
> Code: 04 89 37 83 7c 24 10 00 0f 8f 5a ff ff ff 8b 47 18 2b 45 00 89
> 47 18 83 7d 00 00 0f 85 61 02 00 00 f7 44 24 04 0e 80 f8 ff 74 02 <0f>
> 0b f7 44 24 04 00 20 00 00 0f 85 2d 02 00 00 8b 5c 24 04 81
> EIP: [<b014932e>] cache_alloc_refill+0x18e/0x400 SS:ESP 0068:bd2bbe00
> ide0 at 0x1f0-0x1f7,0x3f6 on irq 14

cu
Adrian

-- 

       "Is there not promise of rain?" Ling Tan asked suddenly out
        of the darkness. There had been need of rain for many days.
       "Only a promise," Lao Er said.
                                       Pearl S. Buck - Dragon Seed

Re: 2.6 git kernel reporting bug in knodemgrd_0 during boot

[Stefan Richter]

Adrian Bunk wrote:
> On Sun, Oct 29, 2006 at 04:41:54PM -0500, Andrew Paprocki wrote:
>> I just upgraded my dev box to the latest 2.6 source via git and this
>> is now printing out in dmesg upon every boot.
...

Andrew, what was the last kernel which didn't log this?

This seems to be either a previously unknown but old problem in the
csr1212 library or a new problem outside of the ieee1394 subsystem.

> The ieee1394 maintainers (Cc'ed) might be the best contact.
...
Thanks Adrian. (I watch lkml but it's easy to miss posts due to lkml's
high traffic.)
-- 
Stefan Richter
-=====-=-==- =-=- ===-=
http://arcgraph.de/sr/

Re: 2.6 git kernel reporting bug in knodemgrd_0 during boot

[Stefan Richter]

I wrote:
>> On Sun, Oct 29, 2006 at 04:41:54PM -0500, Andrew Paprocki wrote:
>>> I just upgraded my dev box to the latest 2.6 source via git and this
>>> is now printing out in dmesg upon every boot.
> ...
> Andrew, what was the last kernel which didn't log this?

And did you change any of the IEEE1394 options in the kernel configuration?
-- 
Stefan Richter
-=====-=-==- =-=- ====-
http://arcgraph.de/sr/

Re: 2.6 git kernel reporting bug in knodemgrd_0 during boot

[Andrew Paprocki]

> > Andrew, what was the last kernel which didn't log this?

I was previously using a stock Debian snapshot from 2006-10-11 which
comes with the 2.6.17 kernel standard (vmlinuz-2.6.17-2-486 is what it
installs).
I had already rebuilt the 2.6.17 source that comes with that install,
tweaking the configuration for the VIA motherboard and it did not
report any problems. Last night I just git'd the latest 2.6 source and
copied the same .config from the previous kernel over, and answered
any new questions that popped up during make config for the newer
source.

> And did you change any of the IEEE1394 options in the kernel configuration?

I copied the .config from the working kernel over, and I don't believe
I changed anything from the working 2.6.17 configuration. These are
the options currently set:

# grep IEEE .config
# CONFIG_IEEE80211 is not set
# IEEE 1394 (FireWire) support
CONFIG_IEEE1394=y
# CONFIG_IEEE1394_VERBOSEDEBUG is not set
CONFIG_IEEE1394_OUI_DB=y
CONFIG_IEEE1394_EXTRA_CONFIG_ROMS=y
CONFIG_IEEE1394_CONFIG_ROM_IP1394=y
# CONFIG_IEEE1394_EXPORT_FULL_API is not set
# CONFIG_IEEE1394_PCILYNX is not set
CONFIG_IEEE1394_OHCI1394=y
# CONFIG_IEEE1394_VIDEO1394 is not set
CONFIG_IEEE1394_SBP2=y
CONFIG_IEEE1394_SBP2_PHYS_DMA=y
# CONFIG_IEEE1394_ETH1394 is not set
# CONFIG_IEEE1394_DV1394 is not set
# CONFIG_IEEE1394_RAWIO is not set

-Andrew