RHEL5: System hangs with softlockup

Issue

• Our RHEL 5 system occasionly locks up and requires a reset to recover. Finding "soft lockup" messages in /var/log/messages:

kernel: Pid: 25217, comm: bpbkar Not tainted 2.6.18-308.11.1.el5 #1
kernel: RIP: 0010:[<ffffffff80016278>]  [<ffffffff80016278>] __bitmap_empty+0xf/0x62
kernel: RSP: 0018:ffff810442e35d48  EFLAGS: 00000212
kernel: RAX: 0000000000000003 RBX: ffff81062b060200 RCX: 00000000000000ff
kernel: RDX: 000000000000003f RSI: 00000000000000ff RDI: ffff81062b060200
kernel: RBP: 0000000000000000 R08: 0000000000000004 R09: ffff81062b060200
kernel: R10: 0000000000000296 R11: 0000000000000000 R12: ffffffff8002b44d
kernel: R13: ffff810442e35cb8 R14: 0000000024503c78 R15: ffff810624569660
kernel: FS:  00002b1ebbf2a080(0000) GS:ffff81062b1e0440(0000) knlGS:00000000f759a9e0
kernel: CS:  0010 DS: 002b ES: 002b CR0: 000000008005003b
kernel: CR2: 00000000080ef678 CR3: 0000000000201000 CR4: 00000000000006e0
kernel:
kernel: Call Trace:
kernel:  [<ffffffff80023185>] flush_tlb_others+0x9a/0xbd
kernel:  [<ffffffff80077afe>] flush_tlb_mm+0xcc/0xd7
kernel:  [<ffffffff80007c8b>] unmap_vmas+0x5b4/0x909
kernel:  [<ffffffff80039e46>] exit_mmap+0x87/0x104
kernel:  [<ffffffff8003bfd5>] mmput+0x30/0x82
kernel:  [<ffffffff800158b4>] do_exit+0x2e7/0x931
kernel:  [<ffffffff80048e4a>] cpuset_exit+0x0/0x88
kernel:  [<ffffffff80061624>] cstar_do_call+0x1b/0x6e

also seeing a hung task timeout like this:

kernel: INFO: task crond:25236 blocked for more than 120 seconds.
kernel: "echo 0 > /proc/sys/kernel/hung_task_timeout_secs" disables this message.
kernel: crond         D ffff81022b061420     0 25236   4399                     (NOTLB)
kernel:  ffff8100287bbe58 0000000000000082 80c40b0607d1142f f60b14fec3c66db6
kernel:  4b493b0b9722722c 0000000000000009 ffff8102225b40c0 ffff81062b0ef080
kernel:  000c503001a97ac6 000000000020ea4c ffff8102225b42a8 00000002801add06
kernel: Call Trace:
kernel:  [<ffffffff80063171>] wait_for_completion+0x79/0xa2
kernel:  [<ffffffff8008ee84>] default_wake_function+0x0/0xe
kernel:  [<ffffffff801296e5>] __key_instantiate_and_link+0x8f/0xc5
kernel:  [<ffffffff800a1820>] synchronize_rcu+0x30/0x36
kernel:  [<ffffffff800a135c>] wakeme_after_rcu+0x0/0x9
kernel:  [<ffffffff8012c105>] install_session_keyring+0xc0/0xd3
kernel:  [<ffffffff8012c633>] join_session_keyring+0x25/0xcb
kernel:  [<ffffffff8012baf0>] keyctl_join_session_keyring+0x2d/0x40
kernel:  [<ffffffff8005d28d>] tracesys+0xd5/0xe0

These are occurring over and over again and seem to be linked.