[OpenAFS-devel] Problems on 8-way Itanium2 system

[Alf Wachsmann]

On Sat, 29 Jan 2005, Derrick J Brashear wrote:
> That's a very strange looking backtrace. In particular, while it's an afs
> client process, nothing in the stack is afs specific at all. What do the
> tracebacks look like for the other processes on the machine? And can you
> reduce still further to 2 processors and reproduce?

I cannot make the machine hang with only 2 CPUs. Reproducing the
problem with 4 CPUs is no problem.

>From the 4 CPU hang, I took an complete crash dump using SGI's kdb
feature. Let me know if you want to look at it.

I have appended a backtrace from this case below. It looks a little
better than the trace I posted before.

Many thanks,
                  Alf.



[2]kdb> bt
Stack traceback for pid 1262
0xe0000034f33c8000     1262        1  1    2   R  0xe0000034f33c85a0 *afs_background
0xe000000004408c20 ia64_spinlock_contention+0x20
        args (0xe0000134f3e3b500, 0x1, 0xe0000134f3e3b774, 0xe0000134f3e3b770, 0xe0000034f33cfc40)
        kernel .text 0xe000000004400000 0xe000000004408c00 0xe000000004408c50
        r31 (spinlock address) 0xe000000004ea4380 kernel_flag
0xe0000000045f03b0 ext2_new_block+0x11d0
        kernel .text 0xe000000004400000 0xe0000000045ef1e0 0xe0000000045f0d80
0xe0000000045f53f0 ext2_alloc_block+0x190
        args (0xe0000134f3e3b500, 0x736d4, 0xe0000034f33cfc40, 0xe0000134f3e3b774, 0xe0000000045f5a50)
        kernel .text 0xe000000004400000 0xe0000000045f5260 0xe0000000045f54a0
0xe0000000045f5a50 ext2_alloc_branch+0x70
        args (0xe0000134f3e3b500, 0x1, 0x736d4, 0xe0000034f33cfc54, 0xe0000034f33cfc78)
        kernel .text 0xe000000004400000 0xe0000000045f59e0 0xe0000000045f5f20
0xe0000000045f63a0 ext2_get_block+0x480
        args (0xe0000134f3e3b500, 0x19, 0xe00000b0083bdd80, 0x1, 0xe0000134f3e3b780)
        kernel .text 0xe000000004400000 0xe0000000045f5f20 0xe0000000045f6ae0
0xe00000000458a030 __block_prepare_write+0x4b0
        args (0xe0000134f3e3b500, 0xa0007ffdb8b4e420, 0x1000, 0x2000, 0xe000000004d436a0)
        kernel .text 0xe000000004400000 0xe000000004589b80 0xe00000000458a400
0xe00000000458b610 block_prepare_write+0x50
        args (0xa0007ffdb8b4e450, 0x1000, 0x2000, 0xe000000004d436a0, 0xe0000000045f6bd0)
        kernel .text 0xe000000004400000 0xe00000000458b5c0 0xe00000000458b680
0xe0000000045f6bd0 ext2_prepare_write+0x30
        args (0xe000003011207450, 0xa0007ffdb8b4e420, 0x1000, 0x2000, 0xe0000000045568c0)
        kernel .text 0xe000000004400000 0xe0000000045f6ba0 0xe0000000045f6c00
0xe0000000045568c0 generic_file_write_nolock+0x700
        args (0xe000003011207450, 0xe0000134f6204000, 0x1000, 0xe000003011207488, 0x301b42c000)
        kernel .text 0xe000000004400000 0xe0000000045561c0 0xe000000004556dc0
0xe000000004556e60 generic_file_write+0xa0
        args (0xe000003011207450, 0xe0000134f6204000, 0x1000, 0xe000003011207488, 0xe0000134f3e3b5c0)
        kernel .text 0xe000000004400000 0xe000000004556dc0 0xe000000004556ec0
0xa00000000987cf50 [libafs-2.4.21-sgi302r24-ia64.mp]osi_rdwr+0x190
        args (0x1, 0xa0000000098cca08, 0xe0000134f6204000, 0x1000, 0xe0000034f33cfd00)
        libafs-2.4.21-sgi302r24-ia64.mp .text 0xa0000000097b80c0 0xa00000000987cdc0 0xa00000000987d000
0xa00000000987c620 [libafs-2.4.21-sgi302r24-ia64.mp]afs_osi_Write+0x120
        args (0xe000003011207380, 0xffffffffffffffff, 0xe0000134f6204000, 0x1000, 0xa0000000098cbf98)
        libafs-2.4.21-sgi302r24-ia64.mp .text 0xa0000000097b80c0 0xa00000000987c500 0xa00000000987c840
0xa0000000097d5110 [libafs-2.4.21-sgi302r24-ia64.mp]afs_UFSCacheFetchProc+0x410
        args (0xe00000b4f3ad6780, 0xe000003011207380, 0xac1a000, 0xa000000009daf920, 0xa0000000099e78a8)
        libafs-2.4.21-sgi302r24-ia64.mp .text 0xa0000000097b80c0 0xa0000000097d4d00 0xa0000000097d52c0
0xa0000000097d81b0 [libafs-2.4.21-sgi302r24-ia64.mp]afs_GetDCache+0x2c70
        args (0xa0000000099e75c0, 0xac00000, 0xe0000034f33cfdd0, 0xe0000034f33cfdf0, 0xe0000034f33cfdf4)
        libafs-2.4.21-sgi302r24-ia64.mp .text 0xa0000000097b80c0 0xa0000000097d5540 0xa0000000097da1a0
0xa0000000097cf010 [libafs-2.4.21-sgi302r24-ia64.mp]BPrefetch+0xf0
        args (0xa0000000098d8950, 0xa0000000098cca08, 0xa0000000098cca08, 0xa0000000097d0170, 0xf22)
        libafs-2.4.21-sgi302r24-ia64.mp .text 0xa0000000097b80c0 0xa0000000097cef20 0xa0000000097cf0e0
0xa0000000097d0170 [libafs-2.4.21-sgi302r24-ia64.mp]afs_BackgroundDaemon+0x6b0
        args (0xa0000000098d8986, 0xa0000000098cca08, 0xa0000000098d8950, 0x0, 0xa0000000098d8c99)
        libafs-2.4.21-sgi302r24-ia64.mp .text 0xa0000000097b80c0 0xa0000000097cfac0 0xa0000000097d0380
0xa000000009889bd0 [libafs-2.4.21-sgi302r24-ia64.mp]afsd_thread+0x6b0
        args (0xe0000034f340fb70, 0xa0000000098cca08, 0xa0000000098cbfb8, 0xe0000034f33c8ec0, 0x0)
        libafs-2.4.21-sgi302r24-ia64.mp .text 0xa0000000097b80c0 0xa000000009889520 0xa00000000988a3e0
0xe000000004416ae0 arch_kernel_thread+0x160
        args (0xa0000000098ce988, 0xe00000000515b200, 0x0, 0xffffffffff841120, 0xe0000000044f1b10)
        kernel .text 0xe000000004400000 0xe000000004416980 0xe000000004416c00
0xe0000000044f1b10 kernel_thread+0xf0
        args (0xa0000000098ce980, 0xe0000034f340fb70, 0x4011, 0x0, 0xa00000000988a410)
        kernel .text 0xe000000004400000 0xe0000000044f1a20 0xe0000000044f1b40
0xa00000000988a410 [libafs-2.4.21-sgi302r24-ia64.mp]afsd_launcher+0x30
        args (0xa0000000098cca08, 0xe000000004507bd0, 0x205)
        libafs-2.4.21-sgi302r24-ia64.mp .text 0xa0000000097b80c0 0xa00000000988a3e0 0xa00000000988a480
0xe000000004507bd0 __run_task_queue+0x1b0
        args (0xe000003016f6fe00, 0xe00000000515b200, 0xe000000004525c90, 0x510)
        kernel .text 0xe000000004400000 0xe000000004507a20 0xe000000004507c20
0xe000000004525c90 context_thread+0x2d0
        args (0xe000000004f7cf60, 0xe000003016f68000, 0xe000000004f7cf70, 0xe000003016f68010, 0xe000000004f7cf88)
        kernel .text 0xe000000004400000 0xe0000000045259c0 0xe000000004525dc0
0xe000000004416ae0 arch_kernel_thread+0x160
        args (0xe000000004d412e8, 0xe00000000515b200, 0x0, 0x24, 0xe0000000044f1b10)
        kernel .text 0xe000000004400000 0xe000000004416980 0xe000000004416c00
0xe0000000044f1b10 kernel_thread+0xf0
        args (0xe000000004d412e0, 0xe000000004e863c8, 0x600, 0x0, 0xe000000004525f30)
        kernel .text 0xe000000004400000 0xe0000000044f1a20 0xe0000000044f1b40
0xe000000004525f30 start_context_thread+0x30
        args (0xe000000004e863c8, 0xe000000004d49110, 0x102)
        kernel .text 0xe000000004400000 0xe000000004525f00 0xe000000004525f60
0xe000000004d49110 do_basic_setup+0x90
        args (0xe000000004408e20, 0x20a)
        kernel .text.init 0xe000000004d48000 0xe000000004d49080 0xe000000004d49140
0xe000000004408e20 init+0xc0
        args (0x0, 0xe0000034f72185a8, 0xe000000004416ae0, 0x30c)
        kernel .text 0xe000000004400000 0xe000000004408d60 0xe0000000044090a0
0xe000000004416ae0 arch_kernel_thread+0x160
        args (0xe000000004d3ab68, 0xe00000000515b200, 0x0, 0xaeeeeeee8badbeef, 0xe0000000044f1b10)
        kernel .text 0xe000000004400000 0xe000000004416980 0xe000000004416c00
0xe0000000044f1b10 kernel_thread+0xf0
        args (0xe000000004d3ab60, 0x0, 0xe00, 0x0, 0xe000000004408cd0)
        kernel .text 0xe000000004400000 0xe0000000044f1a20 0xe0000000044f1b40
0xe000000004408cd0 rest_init+0x50
        args (0xe000000004d48f60, 0x58e)
        kernel .text 0xe000000004400000 0xe000000004408c80 0xe000000004408d60
0xe000000004d48f60 start_kernel+0x480
        args (0x1b4f7da6e08, 0xca6, 0x1300265f700, 0x13002725000, 0x1b4f7d4a7b0)
        kernel .text.init 0xe000000004d48000 0xe000000004d48ae0 0xe000000004d48fe0
0xe0000000044081c0 start_ap+0x2a0
        args (0x1b4f7f76000, 0x300303ebd0, 0x0, 0x1, 0x1b4f7da6e08)
        kernel .text 0xe000000004400000 0xe000000004407f20 0xe0000000044081e0
unwind.unw_unwind: bsp (0xe0000034f33cafc0) out of range [0xe0000034f33cb030-0xe0000034f33cbc18]