[openib-general] OOPS: ib_mad crashery on bootup

[Jeff Carr]

On 05/25/05 10:18, Sean Hefty wrote:
> I've been able to hit an issue in the MAD layer that might be related.  
> I still do not know the root cause, however.

Just to chim in with the same. I've been able to generate a potentially 
similar error on bootup. I put the openib svn on top of 2.6.12-rc4-git7.

Jeff


Configuring network interfaces...Unable to handle kernel NULL pointer 
dereference at virtual address 00000018
  printing eip:
f8a309f2
*pde = 00000000
Oops: 0000 [#1]
PREEMPT SMP
Modules linked in: ib_mthca i2c_i801 i2c_core uhci_hcd usbcore 
pci_hotplug parport_pc parport floppy evdev pcspkr ib_cm ib_ipoib ib_sa 
ib_mad ib_core dm_mod ide_cd cdrom rtc ide_disk ide_generic via82cxxx 
trm290 triflex slc90e66 sis5513 siimage serverworks sc1200 rz1000 
pdc202xx_old pdc202xx_new opti621 ns87415 hpt366 hpt34x generic cy82c693 
cs5530 cs5520 cmd64x atiixp amd74xx alim15x3 aec62xx unix xfs exportfs 
reiserfs ata_piix libata scsi_mod e1000 ext3 jbd mbcache piix ide_core
CPU:    0
EIP:    0060:[<f8a309f2>]    Not tainted VLI
EFLAGS: 00010282   (2.6.12-rc4-git7-20050523)
EIP is at ib_mad_send_done_handler+0x12/0x130 [ib_mad]
eax: f7a73f08   ebx: f74b8680   ecx: 00000001   edx: 00000001
esi: 00000000   edi: df84d874   ebp: f74b8680   esp: f7a73ed0
ds: 007b   es: 007b   ss: 0068
Process ib_mad2 (pid: 2773, threadinfo=f7a72000 task=f7836540)
Stack: df849800 81000001 00000000 00000001 00000001 df84d400 f7a73f08 
df84d800
        df84d874 df84d800 f8a30cec df84d800 f7a73f08 f7a73f08 f74b8680 
00000000
        00000000 00000000 0000000e c2010a14 dfd3a780 00000403 f7a73f8c 
00000000
Call Trace:
  [<f8a30cec>] ib_mad_completion_handler+0x8c/0xb0 [ib_mad]
  [<c012f079>] worker_thread+0x1b9/0x260
  [<f8a30c60>] ib_mad_completion_handler+0x0/0xb0 [ib_mad]
  [<c0119240>] default_wake_function+0x0/0x20
  [<c0119240>] default_wake_function+0x0/0x20
  [<c012eec0>] worker_thread+0x0/0x260
  [<c013386a>] kthread+0xba/0xc0
  [<c01337b0>] kthread+0x0/0xc0
  [<c01014a5>] kernel_thread_helper+0x5/0x10
Code: e8 44 34 00 00 89 44 24 08 48 0f 85 dc fe ff ff e9 28 ff ff ff 8d 
74 26 00 55 57 56 53 83 ec 18 8b 44 24 30 8b 18 8b 73 08 89 dd <8b> 56 
18 89 54 24 10 8d b4 26 00 00 00 00 89 f0 31 ff e8 b7 55
  done.