Created attachment 20922 [details] 2.6.29.1 kernel config Shortly after an upgrade from 2.6.28 to 2.6.29.1, when I was copying a large tree of files and directories, I had an XFS-related kernel error (and machine lockup). Same machine works on 2.6.28 without any problems for weeks. I have rolled back the kernel for now. The filesystem in question is situated on a dm-crypt LUKS volume (if that might've had an impact on something). The error message displayed before lock up was: ----------------------------- invalid opcode: 0000 [#1] SMP last sysfs file: /sys/devices/system/cpu/cpu1/cache/index2/shared_cpu_map Stack: ffff8800bc8e5d60 ffff8800b26015c0 ffff8800bc8e5d60 ffffffffa01b34e6 ffffffff80790d00 ffffffff80790d00 000000000000008a ffffffffa01a445c Call Trace: [<ffffffffa01b4fe6>] ? xfs_inode_set_reclaim_tag+0x86/0xc0 [xfs] [<ffffffffa01a445c>] ? xfs_reclaim+0x4c/0xd0 [xfs] [<ffffffffa01b34e6>] ? xfs_fs_destroy_inode+0x36/0x60 [xfs] [<ffffffff802f45e6>] ? dispose_list+0xa6/0x120 [<ffffffff802f48cf>] ? shrink_icache_memory+0x26f/0x2e0 [<ffffffff802b30ec>] ? shrink_slab+0x12c/0x190 [<ffffffff802b38cb>] ? kswapd+0x4db/0x6c0 [<ffffffff802b0ce0>] ? isolate_pages_global+0x0/0x280 [<ffffffff80261d00>] ? autoremove_wake_function+0x0/0x30 [<ffffffff8023b323>] ? __wake_up_common+0x53/0x80 [<ffffffff802b33f0>] ? kswapd+0x0/0x6c0 [<ffffffff802618c7>] ? kthread+0x47/0x80 [<ffffffff8020d75a>] ? child_rip+0xa/0x20 [<ffffffff80261880>] ? kthread+0x0/0x80 [<ffffffff8020d750>] ? child_rip+0x0/0x20
There's not been any followup, closing.