Kernel crashes

[Goswin von Brederlow <brederlo@informatik.uni-tuebingen.de>]

Hi,

today I found my opteron crashing shortly after boot:
Message from syslogd@opteron at Wed Dec 17 22:38:50 2003 ...
opteron kernel: invalid operand: 0000 [1]

Message from syslogd@opteron at Wed Dec 17 22:38:50 2003 ...
opteron kernel: RIP <ffffffff80121d6c>{__pci_map_cont+108} RSP <000001007ef878e8>

After several tries I managed to start dmesg in a ssh session before
the kernel shutdown completly:

----------- [cut here ] --------- [please bite here ] ---------
Kernel BUG at pci_gart:398
invalid operand: 0000 [1]
CPU 0 
Pid: 69, comm: kjournald Not tainted
RIP: 0010:[<ffffffff80121d6c>] <ffffffff80121d6c>{__pci_map_cont+108}
RSP: 0000:000001007ef878e8  EFLAGS: 00010206
RAX: 0000000000000400 RBX: 0000010008202000 RCX: 000000007f024c00
RDX: 0000000000000000 RSI: 00000000000011b4 RDI: 0000010008202018
RBP: 0000000000000000 R08: 00000000011b4000 R09: 0000000000000001
R10: 000000007ee54400 R11: 00000000000011b3 R12: 0000000000000002
R13: 0000010008202000 R14: 0000000000000002 R15: 00000100081c7000
FS:  0000000000000000(0000) GS:ffffffff804c8b80(0000) knlGS:0000000000000000
CS:  0010 DS: 002b ES: 002b CR0: 000000008005003b
CR2: 00000000556004cc CR3: 0000000000101000 CR4: 00000000000006a0
Process kjournald (pid: 69, stackpage=1007ee685a0)
Stack: 0000010008202030 0000000000000002 0000010008202000 000000007e65d400 
       0000000000000005 ffffffff80121fbf 000001007f642ad0 0000000000000002 
       0000000100000000 0000010003bf9000 
Call Trace:<ffffffff80121fbf>{pci_map_sg+367} <ffffffff802c0da9>{ide_build_dmatable+73} 
       <ffffffff802c11f4>{ide_start_dma+36} <ffffffff802c135b>{__ide_dma_write+43} 
       <ffffffff802b8e59>{__ide_do_rw_disk+1273} <ffffffff802b08ec>{ide_wait_stat+252} 
       <ffffffff802ad53a>{start_request+586} <ffffffff802ad94c>{ide_do_request+940} 
       <ffffffff802804bd>{generic_unplug_device+93} <ffffffff802806c9>{blk_run_queues+201} 
       <ffffffff8017785f>{__wait_on_buffer+175} <ffffffff80137ab0>{autoremove_wake_function+0} 
       <ffffffff8017bf56>{bio_alloc+326} <ffffffff80137ab0>{autoremove_wake_function+0} 
       <ffffffff80281f28>{submit_bio+184} <ffffffff801c54c1>{journal_commit_transaction+3057} 
       <ffffffff8026e087>{vt_console_print+775} <ffffffff8013a71c>{__call_console_drivers+76} 
       <ffffffff8013a89f>{call_console_drivers+255} <ffffffff8013ab90>{printk+576} 
       <ffffffff80134bf0>{thread_return+0} <ffffffff801c89e4>{kjournald+340} 
       <ffffffff80137ab0>{autoremove_wake_function+0} <ffffffff80137ab0>{autoremove_wake_function+0} 
       <ffffffff801c8870>{commit_timeout+0} <ffffffff80112b7f>{child_rip+8} 
       <ffffffff801c8890>{kjournald+0} <ffffffff80112b77>{child_rip+0} 
       

Code: 0f 0b ee 61 38 80 ff ff ff ff 8e 01 66 66 66 90 66 66 66 90 
RIP <ffffffff80121d6c>{__pci_map_cont+108} RSP <000001007ef878e8>
 
----------- [cut here ] --------- [please bite here ] ---------

Anyone know what that is?

The kernel is 2.6.0-test11 and has been working the past few
days. Something I installed must trigger this. Last thing I installed
for the base system was netbase to get netorking started on boot. But
I used "ifup -a" manually before that without problems.

I switched back to 2.4.23 and its running a few minutes now.

MfG
        Goswin