RHEL6: HP systems crash with unexpected NMI when intel_iommu=on iommu=pt kernel parameters are set hpwdt

Issue

• Kernel panic with following traces:

An NMI occurred, but unable to determine source.
Kernel panic - not syncing: An NMI occurred, please see the Integrated Management Log for details.

Pid: 0, comm: swapper Not tainted 2.6.32-358.14.1.el6.x86_64 #1
Call Trace:
 <NMI>  [<ffffffff8150d668>] ? panic+0xa7/0x16f
 [<ffffffffa00404cd>] ? hpwdt_pretimeout+0x8d/0xbc [hpwdt]
 [<ffffffff81513875>] ? notifier_call_chain+0x55/0x80
 [<ffffffff815138da>] ? atomic_notifier_call_chain+0x1a/0x20
 [<ffffffff8109cc3e>] ? notify_die+0x2e/0x30
 [<ffffffff81511597>] ? do_nmi+0x217/0x340
 [<ffffffff81510e00>] ? nmi+0x20/0x30
 [<ffffffff81307665>] ? acpi_idle_do_entry+0x5d/0x67
 <<EOE>>  [<ffffffff81307ac6>] ? acpi_idle_enter_simple+0xc6/0x14b
 [<ffffffff814153a7>] ? cpuidle_idle_call+0xa7/0x140
 [<ffffffff81009fc6>] ? cpu_idle+0xb6/0x110
 [<ffffffff815071bc>] ? start_secondary+0x2ac/0x2ef

• IML(Integrated Management Log) logs are as follows:

An Unrecoverable System Error (NMI) has occurred (System error code 0x00000000, 0x00000000)