bugcheck! message at shutdown
Hi all,
I got bugcheck! message when I shutdown my Tiger4. I noticed this error after I updated the kernel to 2.6.5. Alex posted same kind of mail before.
http://marc.theaimsgroup.com/?l=linux-kernel&m=108127270905214&w=2
So, I sent him mail.
But he doesn't know the proper fix(he just removed __init from the declaration of get_boot_pages to get rid of bugcheck!).
If you have any information, please let me know.
Thanks,
Masao Fukuchi
//
ksymoops 2.4.8 on ia64 2.6.5. Options used
-v vmlinux (specified)
-K (specified)
-l /proc/modules (default)
-o /lib/modules/2.6.5/ (default)
-m System.map (specified)
No modules in ksyms, skipping objects
No ksyms, skipping lsmod
Pid: 2748, CPU 0, comm: halt
psr : 0000101008026018 ifs : 8000000000000002 ip : [<a000000100641c80>] Tain
ted: GF
Using defaults from ksymoops -t elf64-ia64-little -a ia64-elf64 b0 : a0000001002375a0 b6 : a000000100236000 b7 : a000000100237520 f6 : 1003e000000000000000f f7 : 0ffdee000000000000000 f8 : 10002f000000000000000 f9 : 1003e0000000000000001 f10 : 0fffedffffffff2000000 f11 : 1003e0000000000000000 r1 : a0000001008c0000 r2 : 0000000000000309 r3 : a000000100784f20 r8 : a000000100773c08 r9 : a000000100784f20 r10 : 0000000000000001 r11 : 0000000000000300 r12 : e000000002377c80 r13 : e000000002370000 r14 : a000000100788460 r15 : e000000002370ea4 r16 : a0000001006dd398 r17 : 0000000000000000 r18 : ffffffffffff00e8 r19 : a0000001006a8100 r20 : 0000000000000003 r21 : a0000001006dd378 r22 : a0000001006dd370 r23 : e0000000010072d8 r24 : a0000001007d7a00 r25 : e0000000010072e0 r26 : 0000000000000001 r27 : 0000000000000001 r28 : a000000100099a30 r29 : a000000100785a70 r30 : a000000100788420 r31 : 0000000000000004
Call Trace: [<a000000100019760>] show_stack+0x80/0xa0 [<a00000010003fca0>] die+0x1e0/0x2a0 [<a00000010003fff0>] ia64_bad_break+0x230/0x340 [<a000000100011ea0>] ia64_leave_kernel+0x0/0x260 [<a000000100641c80>] get_boot_pages+0x0/0x300 [<a0000001002375a0>] swiotlb_alloc_coherent+0x80/0x180 [<a0000002000c0330>] mptscsih_synchronize_cache+0x690/0x760 [mptscsih] [<a0000001002375a0>] swiotlb_alloc_coherent+0x80/0x180
Warning (Oops_read): Code line not seen, dumping what data is available
Quoted 16 lines
Quoted 16 lines
IP;
a000000100641c80<get_boot_pages+0/300> <=====b0; a0000001002375a0 <swiotlb_alloc_coherent+80/180> b6; a000000100236000 <__copy_user+5e0/920> b7; a000000100237520 <swiotlb_alloc_coherent+0/180>r1;
a0000001008c0000<__gp+0/3ffffffeff740000> r3;a000000100784f20<system_running+0/8> r8;a000000100773c08<kallsyms_names+222e8/31ab0> r9;a000000100784f20<system_running+0/8> r12;e000000002377c80<v+2377c80/1fffffffffff0000> r13;e000000002370000<v+2370000/1fffffffffff0000> r14;a000000100788460<ia64_mv+40/160> r15;e000000002370ea4<v+2370ea4/1fffffffffff0000> r16;a0000001006dd398<console_sem+0/20> r18;ffffffffffff00e8<per_cpu__cpu_info+c0/c8> r19;a0000001006a8100<cpu_to_node_map+0/80>
Quoted 8 lines
Quoted 8 lines
r21;
a0000001006dd378<log_wait+8/18>
r22;a0000001006dd370<log_wait+0/18>
r23;e0000000010072d8<v+10072d8/1fffffffffff0000> r24;a0000001007d7a00<console_timer+8/40> r25;e0000000010072e0<v+10072e0/1fffffffffff0000> r28;a000000100099a30<release_console_sem+50/220> r29;a000000100785a70<blank_state+0/4> r30;a000000100788420<ia64_mv+0/160>
Trace; a000000100019760 <show_stack+80/a0> Trace; a00000010003fca0 <die+1e0/2a0>
Trace; a00000010003fff0 <ia64_bad_break+230/340> Trace; a000000100011ea0 <ia64_leave_kernel+0/260> Trace; a000000100641c80 <get_boot_pages+0/300> Trace; a0000001002375a0 <swiotlb_alloc_coherent+80/180> Trace; a0000002000c0330 <__gp+ff800330/3ffffffeff740000> Trace; a0000001002375a0 <swiotlb_alloc_coherent+80/180>
Alex Williamson wrote:
Hi Masao,
You can remove the __init from the declaration of get_boot_pages, but
I'm not sure that's the proper fix. It gets rid of the bugcheck for me though. Please post to linux-kernel as well, perhaps it will get someone's attention who knows how to fix it properly. I'd assume we really only want to call get_boot_pages on bootup, and the system_running flag is being used improperly. Thanks,
Alex