BUG: soft lockup detected on CPU#0!

Problem

When performance very large I/O transfer using LSI MegaRAID 320-1 or 320-1LP controller, the kernel dumps the following errors in syslog:

BUG: soft lockup detected on CPU#0!
CPU 0:
Modules linked in: ipv6 autofs4 sunrpc dm_mod video button battery ac ohci_hcd i2c_amd8111 i2c_cor
e hw_random k8_edac edac_mc tg3 floppy ext3 jbd megaraid_mbox megaraid_mm sd_mod scsi_mod
Pid: 26, comm: kblockd/0 Not tainted 2.6.16-1.2096_FC3asl.6smp #1
RIP: 0010:[] {_spin_unlock_irq+10}
RSP: 0000:ffff810209b55de0  EFLAGS: 00000246
RAX: 0000000000000000 RBX: 0000000000000000 RCX: 0000000000000247
RDX: 0000000000000247 RSI: 0000000000000000 RDI: ffff8101ff0b2248
RBP: ffff8101ff001800 R08: 0000000000000000 R09: 0000000300000000
R10: 0000000300000000 R11: 0000000000000046 R12: ffff8101ff808800
R13: ffff8101ff0b2048 R14: ffff8101ff0b2048 R15: ffffffff801fed4e
FS:  00002aaaaadf2b00(0000) GS:ffffffff80512000(0000) knlGS:0000000000000000
CS:  0010 DS: 0018 ES: 0018 CR0: 000000008005003b
CR2: 00007fffffd66ba0 CR3: 00000001fdac7000 CR4: 00000000000006e0

Call Trace: {:scsi_mod:scsi_request_fn+818}
       {cfq_kick_queue+127} {run_workqueue+161}
       {worker_thread+0} {keventd_create_kthread+0}
       {worker_thread+261} {default_wake_function+0}
       {keventd_create_kthread+0} {default_wake_function+0}
       {keventd_create_kthread+0} {kthread+263}
       {child_rip+8} {keventd_create_kthread+0}
       {kthread+0} {child_rip+0}

This symptom has exhibited under 2.6.12-X_FCXaslmp and 2.6.16-1.2096_FCXasl.Xsmp kernels.

Solution

This is kernel bug that has been fixed in the newer 2.6.17-1.2157_FCXasl.Xsmp kernel.

Support

BUG: soft lockup detected on CPU#0!

Problem

Solution