IBM Support

CPU DLPAR add or remove operations may fail with Broadcom Emulex Fibre Channel adapters and cause system crash

Flashes (Alerts)


Abstract

Power Systems with Broadcom Emulex Fibre Channel adapters might fail to remove and add CPUs by using DLPAR operations. This failure can be seen in Red Hat Enterprise Linux 8 and SUSE Linux Enterprise Server 15.

Content

Linux Releases Affected
Red Hat Enterprise Linux (RHEL) 8.2
Red Hat Enterprise Linux (RHEL) 8.3
SLES 15
IBM Systems Affected
All IBM Power Systems with Emulex FC adapters.
Symptoms
When performing DLPAR operations on CPUs while the FC adapter is installed, there is a possibility 
that the driver will not register addition of new CPUs or removal of active CPUs. This can 
cause the system to crash. The crash may look similar to the following call trace:
 [c0000001d47bb5a0] [d0000000041daddc] lpfc_sli4_poll_hbtimer+0xc4/0xe0 [lpfc]
 [c0000001d47bb5e0] [c000000000227150] call_timer_fn+0x50/0x200
 [c0000001d47bb670] [c000000000227444] expire_timers+0x144/0x220
 [c0000001d47bb6e0] [c000000000227c30] run_timer_softirq+0x3f0/0xe80
 [c0000001d47bb7b0] [c000000000ddeff8] __do_softirq+0x178/0x420
 [c0000001d47bb8b0] [c00000000015e884] irq_exit+0x184/0x1c0
 [c0000001d47bb8e0] [c00000000002a888] timer_interrupt+0x128/0x2f0
 [c0000001d47bb940] [c0000000000091c4] decrementer_common+0x114/0x120
 [c0000001d47bbc30] [c0000001d47bbca0] 0xc0000001d47bbca0 (unreliable)
 [c0000001d47bbca0] [c00000000029158c] cpu_stopper_thread+0x14c/0x240
 [c0000001d47bbd50] [c000000000197e48] smpboot_thread_fn+0x1e8/0x2a0
 [c0000001d47bbdc0] [c00000000019005c] kthread+0x1ac/0x1c0
 [c0000001d47bbe30] [c00000000000b75c] ret_from_kernel_thread+0x5c/0x80
Workaround

There is no workaround for this issue. However, rather than using the DLPAR operation it is recommended to shut down the logical partition before removing or adding CPUs to the configuration.

For more information on DLPAR, see Dynamic logical partitioning.

Fix Outlook
The resolution to this issue is in a patch that was applied to the following releases of RHEL and SLES.

RHEL 8.4 and newer versions of RHEL 8
SLES 15sp3 kernel version 5.3.18-24.9 or later

Upgrading the kernel to one of these later versions must resolve the issue. If the issue is still seen, reach out to IBM support for further assistance.

[{"Type":"MASTER","Line of Business":{"code":"LOB26","label":"Storage"},"Business Unit":{"code":"BU058","label":"IBM Infrastructure w\/TPS"},"Product":{"code":"SGMV157","label":"IBM Support for Red Hat Enterprise Linux Server"},"ARM Category":[{"code":"a8m0z000000Gnl7AAC","label":"Red Hat Enterprise Linux"}],"Platform":[{"code":"PF025","label":"Platform Independent"}],"Version":"All Versions"}]

Document Information

Modified date:
15 August 2022

UID

ibm16610833