Installation of SLES10 SP2 on NEC ia64 machine fails with NULL pointer dereference and IP at __migrate_task

This document (7003509) is provided subject to the disclaimer at the end of this document.

Environment

SUSE Linux Enterprise Server 10 Service Pack 2

Situation

Installation of SLES10 SP2 on a NEC ia64 machine with 32 CPUs and 128 GB RAM fails with the oops/panic that looks like the following:

Boot processor id 0x0/0x0
Unable to handle kernel NULL pointer dereference (address 00000000000012f3)
migration/0[2]: Oops 11012296146944 [1]
Modules linked in:

Pid: 2, CPU 0, comm:          migration/0
psr : 00001210085a2010 ifs : 800000000000050d ip  : [<a000000100085e41>]    Not
tainted
ip is at __migrate_task+0xc1/0x220
unat: 0000000000000000 pfs : 000000000000050d rsc : 0000000000000003
rnat: 0000000000000000 bsps: 0000000000000000 pr  : 0000000000009581
ldrs: 0000000000000000 ccv : 0000000000000000 fpsr: 0009804c8a70433f
csd : 893fffff000f0000 ssd : 893fffff00000000
b0  : a000000100085e30 b6  : a000000100090140 b7  : a00000010000c3e0
f6  : 0fffefffffffff0000000 f7  : 0ffe9886c000000000000
f8  : 1000d886c000000000000 f9  : 0ffff8000000000000000
f10 : 1000d886bfffff7794000 f11 : 1003e0000000000004436
r1  : a0000001011dd210 r2  : 0000000000000286 r3  : 0000000000000001
r8  : a000000101279158 r9  : a0000001010feeb0 r10 : 0000000000000001
r11 : a000000101279158 r12 : e000000135027a50 r13 : e000000135020000
r14 : 0000000000000000 r15 : 0000000000000107 r16 : a0000001010feeb0
r17 : a0000001010feeb0 r18 : 00000000000012f3 r19 : a000000101279158
r20 : 0000000000000000 r21 : e000000001110000 r22 : e000000001110000
r23 : ffffffffffff67e0 r24 : 0000000000008040 r25 : ffffffffffff67d8
r26 : ffffffffffff67d8 r27 : 0000000000000000 r28 : ffffffffffff67d0
r29 : 0000000000000000 r30 : 0000000000000000 r31 : e000000001105580

Call Trace:
 [<a0000001000152d0>] show_stack+0x50/0xa0
                                sp=e0000001350275e0 bsp=e000000135021500
 [<a000000100015be0>] show_regs+0x860/0x8a0
                sp=e0000001350277b0 bsp=e0000001350214a8
 [<a00000010003ae30>] die+0x1d0/0x360
                                sp=e0000001350277b0 bsp=e000000135021460
 [<a0000001005a9a00>] ia64_do_page_fault+0xa40/0xbc0
                                sp=e0000001350277d0 bsp=e000000135021410
 [<a00000010000cbe0>] ia64_leave_kernel+0x0/0x280
                                sp=e000000135027880 bsp=e000000135021410
 [<a000000100085e40>] __migrate_task+0xc0/0x220
                                sp=e000000135027a50 bsp=e0000001350213a0
 [<a000000100090650>] migration_thread+0x510/0x5e0
                                sp=e000000135027a50 bsp=e000000135021350
 [<a0000001000d39e0>] kthread+0x240/0x2c0
                                sp=e000000135027a50 bsp=e000000135021308
 [<a0000001000133f0>] kernel_thread_helper+0xd0/0x100
                                sp=e000000135027e30 bsp=e0000001350212e0
 [<a0000001000094c0>] start_kernel_thread+0x20/0x40
                                sp=e000000135027e30 bsp=e0000001350212e0

Entering kdb (current=0xe000000135020000, pid 2) on processor 0 Oops: <NULL>
due to oops @ 0xa000000100085e41
 psr: 0x00001210085a2010   ifs: 0x800000000000050d    ip: 0xa000000100085e40
unat: 0x0000000000000000   pfs: 0x000000000000050d   rsc: 0x0000000000000003
rnat: 0x0000000000000000  bsps: 0x0000000000000000    pr: 0x0000000000009581
ldrs: 0x0000000000000000   ccv: 0x0000000000000000  fpsr: 0x0009804c8a70433f
  b0: 0xa000000100085e30    b6: 0xa000000100090140    b7: 0xa00000010000c3e0
  r1: 0xa0000001011dd210    r2: 0x0000000000000286    r3: 0x0000000000000001
  r8: 0xa000000101279158    r9: 0xa0000001010feeb0   r10: 0x0000000000000001
 r11: 0xa000000101279158   r12: 0xe000000135027a50   r13: 0xe000000135020000
 r14: 0x0000000000000000   r15: 0x0000000000000107   r16: 0xa0000001010feeb0
 r17: 0xa0000001010feeb0   r18: 0x00000000000012f3   r19: 0xa000000101279158
 r20: 0x0000000000000000   r21: 0xe000000001110000   r22: 0xe000000001110000
 r23: 0xffffffffffff67e0   r24: 0x0000000000008040   r25: 0xffffffffffff67d8
 r26: 0xffffffffffff67d8   r27: 0x0000000000000000   r28: 0xffffffffffff67d0
 r29: 0x0000000000000000   r30: 0x0000000000000000   r31: 0xe000000001105580
&regs = e000000135027890
Forcing dump (if configured)

Resolution

A maintenance update of the kernel has been released. Please use kernel version 2.6.16.60-042.4 or later. In case you need an ISO image for installation please contact SUSE Technical Services.

Additional Information

This problem is due to a faulty patch that has been introduced in SLES10 SP2.

Disclaimer

This Support Knowledgebase provides a valuable tool for SUSE customers and parties interested in our products and solutions to acquire information, ideas and learn from one another. Materials are provided for informational, personal or non-commercial use within your organization and are presented "AS IS" WITHOUT WARRANTY OF ANY KIND.