[LU-2940] Oops in vvp_pgcache_show()

• Type: Bug
• Resolution: Fixed
• Priority: Major
• Fix Version/s: Lustre 2.5.0
• Affects Version/s: Lustre 2.4.0, Lustre 2.5.0
• Labels: llite patch procfs
• Severity: 3
• Assignee: Emoly Liu
• Reporter: John Hammond

Description

To reproduce start racer and run

while true; do cat /proc/fs/lustre/llite/*/dump_page_cache > /dev/null; done

crash> bt -l
PID: 8892   TASK: ffff880118dc8080  CPU: 1   COMMAND: "cat"
 #0 [ffff88013a59b920] machine_kexec at ffffffff81031f7b
    /usr/src/debug/kernel-2.6.32-279.19.1.el6/linux-2.6.32-279.19.1.el6_lustre_gcov.x86_64/arch/x86/kernel/machine_kexec_64.c: 336
 #1 [ffff88013a59b980] crash_kexec at ffffffff810b8c22
    /usr/src/debug/kernel-2.6.32-279.19.1.el6/linux-2.6.32-279.19.1.el6_lustre_gcov.x86_64/kernel/kexec.c: 1106
 #2 [ffff88013a59ba50] oops_end at ffffffff814eef80
    /usr/src/debug/kernel-2.6.32-279.19.1.el6/linux-2.6.32-279.19.1.el6_lustre_gcov.x86_64/arch/x86/kernel/dumpstack.c: 242
 #3 [ffff88013a59ba80] no_context at ffffffff81042a0b
    /usr/src/debug/kernel-2.6.32-279.19.1.el6/linux-2.6.32-279.19.1.el6_lustre_gcov.x86_64/arch/x86/mm/fault.c: 698
 #4 [ffff88013a59bad0] __bad_area_nosemaphore at ffffffff81042c95
    /usr/src/debug/kernel-2.6.32-279.19.1.el6/linux-2.6.32-279.19.1.el6_lustre_gcov.x86_64/arch/x86/mm/fault.c: 764
 #5 [ffff88013a59bb20] bad_area at ffffffff81042dbe
    /usr/src/debug/kernel-2.6.32-279.19.1.el6/linux-2.6.32-279.19.1.el6_lustre_gcov.x86_64/arch/x86/mm/fault.c: 793
 #6 [ffff88013a59bb50] __do_page_fault at ffffffff81043570
    /usr/src/debug/kernel-2.6.32-279.19.1.el6/linux-2.6.32-279.19.1.el6_lustre_gcov.x86_64/arch/x86/mm/fault.c: 1140
 #7 [ffff88013a59bc70] do_page_fault at ffffffff814f0f5e
    /usr/src/debug/kernel-2.6.32-279.19.1.el6/linux-2.6.32-279.19.1.el6_lustre_gcov.x86_64/arch/x86/mm/fault.c: 1208
 #8 [ffff88013a59bca0] page_fault at ffffffff814ee315
    /usr/src/debug/kernel-2.6.32-279.19.1.el6/linux-2.6.32-279.19.1.el6_lustre_gcov.x86_64/arch/x86_64/kernel/entry.S
    [exception RIP: vvp_pgcache_show+263]
    RIP: ffffffffa0df5ab7  RSP: ffff88013a59bd58  RFLAGS: 00010246
    RAX: ffff88010f064ac0  RBX: ffffea00039cdd28  RCX: 0000000000000000
    RDX: 0000000000000000  RSI: ffffffffa0e02b12  RDI: ffff88010f064a00
    RBP: ffff88013a59be18   R8: 00000000fffffffb   R9: 00000000000035bb
    R10: 0000000000000002  R11: 000000000000000f  R12: ffff880132b22440
    R13: ffff88015b589db8  R14: ffff88014ff48948  R15: ffff88010f064a00
    ORIG_RAX: ffffffffffffffff  CS: 0010  SS: 0018
 #9 [ffff88013a59be20] seq_read at ffffffff8119a0a9
    /usr/src/debug/kernel-2.6.32-279.19.1.el6/linux-2.6.32-279.19.1.el6_lustre_gcov.x86_64/fs/seq_file.c: 235
#10 [ffff88013a59bea0] proc_reg_read at ffffffff811dcbde
    /usr/src/debug/kernel-2.6.32-279.19.1.el6/linux-2.6.32-279.19.1.el6_lustre_gcov.x86_64/fs/proc/inode.c: 202
#11 [ffff88013a59bef0] vfs_read at ffffffff811782b5
    /usr/src/debug/kernel-2.6.32-279.19.1.el6/linux-2.6.32-279.19.1.el6_lustre_gcov.x86_64/fs/read_write.c: 295
#12 [ffff88013a59bf30] sys_read at ffffffff811783f1
    /usr/src/debug/kernel-2.6.32-279.19.1.el6/linux-2.6.32-279.19.1.el6_lustre_gcov.x86_64/fs/read_write.c: 369
#13 [ffff88013a59bf80] system_call_fastpath at ffffffff8100b072
    /usr/src/debug///////kernel-2.6.32-279.19.1.el6/linux-2.6.32-279.19.1.el6_lustre_gcov.x86_64/arch/x86/kernel/entry_64.S: 489
    RIP: 0000003282eda360  RSP: 00007fffaf2f4050  RFLAGS: 00010206
    RAX: 0000000000000000  RBX: ffffffff8100b072  RCX: 0000000000b68030
    RDX: 0000000000008000  RSI: 0000000000b60000  RDI: 0000000000000003
    RBP: 0000000000b60000   R8: 000000328318dee8   R9: 0000000000000001
    R10: 0000000000008fff  R11: 0000000000000246  R12: ffffffffffff8000
    R13: 0000000000000003  R14: 0000000000008000  R15: 0000000000000003
    ORIG_RAX: 0000000000000000  CS: 0033  SS: 002b
crash> dis -l vvp_pgcache_show+263
/root/lustre-release/lustre/llite/vvp_dev.c: 420
0xffffffffa0df5ab7 <vvp_pgcache_show+263>:      mov    (%rdx),%rcx