Power8 System Firmware

New features and functions

• On systems with PowerVM firmware, support was added to allow V9R910 and later HMC levels to query Live Partition Mobility (LPM) performance data after an LPM operation.
• Support was added to the Advanced System Management Interface (ASMI) to provide customer control over speculative execution  in response to CVE-2017-5753 and CVE-2017-5715 (collectively known as Spectre) and CVE-2017-5754 (known as Meltdown).   The ASMI "System Configuration/Speculative Execution Control" provides two options that can only be set when the system is powered off:
  1) Speculative execution controls to mitigate user-to-kernel and user-to-user side-channel attacks.  This mode is designed for systems that need to mitigate exposures of the hypervisor, operating systems, and user application data to untrusted code.   This mode is set as the default.
  2) Speculative execution fully enabled:  This optional mode is designed for systems where the hypervisor, operating system, and applications can be fully trusted.
  Note:  Enabling this option could expose the system to CVE-2017-5753, CVE-2017- 5715, and CVE-2017-5754.  This includes any partitions that are migrated (using Live Partition Mobility) to this system.
• On systems with PowerVM firmware, support was added to allow a periodic data capture from the PCIe3 I/O expansion drawer (with feature code #EMX0) cable card links.
• On systems with PowerVM firmware and an IBM i partition,  support was added for multipliers for IBM i MATMATR fields that are limited to four characters.  When retrieving Server metrics via IBM MATMATR calls, and the system contains greater than 9999 GB, for example, MATMATR has an architected "multiplier" field such that 10,000 GB can be represented by 5,000 GB * Multiplier of 2, so '5000' and '2' are returned  in the quantity and multiplier fields, respectively, to handle these extended values.  The IBM i OS also requires a PTF to support the MATMATR field multipliers.

System firmware changes that affect all systems

• A problem was fixed in which deconfigured-resource records can become malformed and cause the loss of service processor for both redundant and non-redundant service processor systems.  These failures can occur during or after firmware updates to the FW860.40, FW860.41, or FW860.42 levels.  The complete loss of service processor results in the loss of HMC (or FSP stand-alone) management of the server and loss of any further error logging.  The server itself will continue to run.  Without the fix, the loss of the service processor could happen within one month of the deconfiguration records being encountered.  It is highly recommended to install the fix.  Recovery from the problem, once encountered, requires a full server AC power cycle and clearing of deconfiguration records to avoid reoccurrence.  Clearing deconfiguration records exposes the server to repeat hardware failures and possible unplanned outages.
• A problem was fixed for the guard reminder processing of garded FRUs and error logs that can cause a system power off to hang and time out with a  service processor reset.
  
• A problem was fixed for the wrong Redfish method (PATCH or POST) passed for a valid Uniform Resource Indicator (URI) causing an incorrect error message of " 501 - Not Implemented".  With the fix, the message returned is "Invalid Method on URI" which is more helpful to the user.
• A problem was fixed for SRC call home reminders for bad FRUs causing service processor dumps with SRC B181E911 and reset/reloads.  This occurred if the FRU callout was missing a CCIN number in the error log.  This can happen because some error logs only have have "Symbolic FRUs" and these were not being handled correctly.

System firmware changes that affect certain systems

• DEFERRED: On systems with PowerVM firmware, a problem was fixed for a PCIe3 I/O expansion drawer (with feature code #EMX0) where control path stability issues may cause certain SRCs to be logged.  Systems using copper cables may log SRC B7006A87 or similar SRCs, and the fanout module may fail to become active.  Systems using optical cables may log SRC of B7006A22 or similar SRCs.  For this problem, the errant I/O drawer may be recovered by a re-IPL of the system.
  
• On systems with PowerVM firmware, a problem was fixed for a Coherent Accelerator Processor Proxy (CAPP) unit hardware failure that caused a hypervisor hang with SRC B7000602.  This failure is very rare and can only occur during the early IPL of the hypervisor, before any partitions are started.   A re-IPL will recover from the problem.
• On systems with PowerVM firmware, a problem was fixed for a Live Partition Mobility migration hang that could occur if one of its VIOS Mover Service Partitions (MSPs) goes into a failover at the start of the LPM operation.  This problem is rare because it requires a MSP error to force a MSP failover at the very start of the LPM migration to get the LPM timing error.  The LPM hang can be recovered by using the "migrlpar -o s" and "migrlpar -o r" commands on the HMC.
• On systems with PowerVM firmware, a problem was fixed for incorrect low affinity scores for a partition reported from the HMC "lsmemopt" command when a partition has filled an entire drawer.  A low score indicates the placement is poor but in this case the placement is actually good.  More information on affinity scores for partitions and the Dynamic Platform Optimizer can be found at the IBM Knowledge Center:  https://www.ibm.com/support/knowledgecenter/en/POWER8/p8hat/p8hat_dpoovw.htm.
• On systems with PowerVM firmware, a problem was fixed to allow the management console to display the Active Memory Mirroring (AMM) licensed capability.  Without the fix, the AMM licensed capability of a server will always show as "off" on the management console, even when it is present.
• On systems with PowerVM firmware, a problem was fixed for a rare hypervisor hang for systems with shared processors with a sharing mode of uncapped.  If this hang occurs, all partitions of the system will become unresponsive and the HMC will go to an "Incomplete" state.
• On systems with PowerVM firmware, a problem was fixed for a Live Partition Mobility migration abort that could occur if one of its VIOS Mover Service Partitions (MSPs) goes into a failover during the LPM operation.  This problem is rare because it requires a MSP error to force a MSP failover during the LPM migration to get the LPM timing error.  The LPM abort can be recovered by retrying the LPM migration.
• On systems with PowerVM firmware and a shared processor pool, a very rare problem was fixed for the hypervisor not responding to partition requests such as power off and LIve Partiton Mobility (LPM).  This error is caused by a request for a guard of a failed processor (when there are not any available spare processors) that has hung.
• On systems using PowerVM firmware with mirrored memory running IBM i partitions, a problem was fixed for un-mirrored nodal memory errors in the partition that also caused the system to crash.   With the fix, the memory failure is isolated to the impacted partition, leaving the rest of the system unaffected.  This fix improves on an earlier fix delivered for  IBM i memory errors  in FW840.60 by handling the errors in nodal memory.
• On systems with PowerVM firmware and Huge Page (16 GB) memory enabled for a AIX partition,  a problem was fixed for the OS failing to boot with an 0607 SRC displayed.  This error occurs on systems with  FW860.40, FW860.41 or FW860.42 installed.  To circumvent the problem, disable Huge Pages for the AIX  partition.  For information on viewing and setting values for AIX huge-page memory allocation, see the following link in the IBM Knowledge Center: https://www.ibm.com/support/knowledgecenter/en/POWER8/p8hat/p8hat_aixviewhgpgmem.htm
• On systems with PowerVM firmware and an IBM i partition, a problem was fixed for 64 bytes overwritten in a portion of the  IBM i Main Storage Dump (MSD).  Approximately 64 bytes are overwritten just beyond the 17 MB (0x11000000) address on P8 systems.  This problem is cosmetic as the dump is still readable for problem diagnostics and no customer operations are affected by it.
• On systems with PowerVM firmware and a partition with a Fibre Channel Adapter (FCA) or a Fibre Channel over Ethernet (FCoE) adapter,  a problem was fixed for bootable disks attached to the FCA/FCoE adapter not being seen in the System Management Services (SMS) menus for selection as boot devices.  This problem is likely to occur if the only I/O device in the partition is a FCA or FCoE adapter.  If other I/O devices are present, the problem may still occur if the FCA or FCoE  is the first adapter discovered by SMS.  A work-around to this problem is to define a virtual Ethernet adapter in the partition profile.  The virtual adapter does not need to have any physical backing device,  as just having the VLAN defined is sufficient to avoid the problem.  The FCA has feature codes #EN0A, #EN0B, #EN0F, #EN0G, #EN0Y, #EN12, #5729, #5774, #5735, and #5723 and the FCoE adapter has feature codes #5708, #EN0H, #EN0J, #EN0K, and #EN0L for all but the Linux on Power  8247 models.  For the Linux on Power 8247 models, the FCA has feature codes  #5729, #5774, #EL43, #EL58, #EL5B, #EL54, and #EL52 and the FCoE adapter has feature codes #5708, #EL56, #EL38, #EL57, and #EL3C.
• On systems with PowerVM firmware and a partition with a 3.0 USB controller, a problem was fixed for a partition boot failure  The USB 3.0 controller may be integrated or a adapter card with feature code #EC45 or #EC46.  The boot failure is triggered by a fault in the USB controller but instead of the just the USB controller failing, the entire partition fails.  With the fix, the failure is limited to the USB controller.
• On systems with PowerVM firmware, a problem was fixed for the FRU callouts for the BA188001 and BA188002 EEH errors to include the PCI Host Bridge (PHB) FRU which had been excluded.  For the P8 systems, these rare errors will more typically isolate to the processor instead of the adapter or slot planar.   In the pre-P8 systems, the I/O planar also included the PHB, but for P8 systems, the PHB was moved to the processor complex.
• On systems using PowerVM firmware, a  problem was fixed for an internal error in the SR-IOV adapter firmware that resets the adapter  and logs a B400FF01 reference code.  This error happens in rare cases when there are multiple partitions actively running traffic through the adapter and a subset of the partitions are shutdown hard.  The error causes a temporary disruption of traffic but recovery from the error is automatic with no user intervention needed.
  This fix updates adapter firmware to 10.2.252.1931, for the following Feature Codes: EN15, EN16, EN17, EN18, EN0H, EN0J, EN0M, EN0N, EN0K, EN0L, EL38, EL3C, EL56, and EL57.
  The SR-IOV adapter firmware level update for the shared-mode adapters happens under user control to prevent unexpected temporary outages on the adapters.  A system reboot will update all SR-IOV shared-mode adapters with the new firmware level.  In addition, when an adapter is first set to SR-IOV shared mode, the adapter firmware is updated to the latest level available with the system firmware (and it is also updated automatically during maintenance operations, such as when the adapter is stopped or replaced).  And lastly, selective manual updates of the SR-IOV adapters can be performed using the Hardware Management Console (HMC).  To selectively update the adapter firmware, follow the steps given at the IBM Knowledge Center for using HMC to make the updates:   https://www.ibm.com/support/knowledgecenter/HW4M4/p8efd/p8efd_updating_sriov_firmware.htm.
  Note: Adapters that are capable of running in SR-IOV mode, but are currently running in dedicated mode and assigned to a partition, can be updated concurrently either by the OS that owns the adapter or the managing HMC (if OS is AIX or VIOS and RMC is running).
  
• On systems using OPAL firmware, Skiboot was updated to V5.4.9 from V5.4.8,  providing the following fix:
  -  A problem was fixed for a possible incorrect value for processor frequency from /proc/cpuinfo. The value being returned was the last frequency requested by the kernel, but may not reflect the current frequency of the processor.
• On systems with PowerVM firmware, a problem was fixed for  a  PCIe3 I/O expansion drawer (with feature code #EMX0)  failing to initialize during the IPL with a SRC B7006A88 logged.  The error is infrequent.  The errant I/O drawer can be recovered by a re-IPL of the system.
• On systems with PowerVM firmware, a problem was fixed for the SR-IOV firmware adapter updates using the HMC GUI or CLI to only reboot one SR-IOV adapter at a time.  If multiple adapters are updated at the same time, the HMC error message HSCF0241E may occur:  "HSCF0241E Could not read firmware information from SR-IOV device ...".  This fix prevents the system network from being disrupted by the SR-IOV adapter updates when redundant configurations are being used for the network.  The problem can be circumvented by using the HMC GUI to update the SR-IOV firmware one adapter at a time using the following steps:  https://www.ibm.com/support/knowledgecenter/en/8247-22L/p8efd/p8efd_updating_sriov_firmware.htm
  

New features and functions

• In response to recently reported security vulnerabilities, this firmware update is being released to address Common Vulnerabilities and Exposures issue numbers CVE-2017-5715, CVE-2017-5753 and CVE-2017-5754.  Operating System updates are required in conjunction with this FW level for CVE-2017-5753 and CVE-2017-5754.
  

System firmware changes that affect certain systems

• DEFERRED:  On systems using PowerVM firmware, a problem was fixed for DPO (Dynamic Platform Optimizer) operations taking a very long and impacting the server system with a performance degradation.  The problem is triggered by a DPO operation being done on a system with unlicensed processor cores and a very high I/O load.  The fix involves using a different lock type for the memory relocation activities (to prevent lock contention between memory relocation threads and partition threads) that is created at IPL time, so an IPL is needed to activate the fix.  More information on the DPO function can be found at the IBM Knowledge Center:  https://www.ibm.com/support/knowledgecenter/en/8247-42L/p8hat/p8hat_dpoovw.htm

System firmware changes that affect certain systems

• DEFERRED: On systems using PowerVM firmware, a problem was fixed for PCIe3 I/O expansion drawer (#EMX0) link improved stability.  The settings for the continuous time linear equalizers (CTLE) was updated for all the PCIe adapters for the PCIe links to the expansion drawer.  The system must be re-IPLed for the fix to activate.

System firmware changes that affect certain systems

• DEFERRED: A problem was fixed for a Field Core Override (FCO) error that causes a processor chip without functional cores to be guarded with a SRC B111BA24 error logged and by guard association causes all the memory and I/O resources behind the processor chip to be lost for the current IPL.  This problem is triggered by a system being  manufactured with one or more feature codes of #2319 (Factory Deconfiguration of 1-core) to assist with optimization of software licensing.  For more information on Field Core Override, refer to IBM Knowledge Center: http://www.ibm.com/support/knowledgecenter/POWER8/p8hby/fieldcore.htm.  The error only occurs in systems where the total number of active cores is less than the number of processor chips.  When the fix is applied on a system that has lost memory or I/O resources due to the errant processor guard, the system must be re-IPLed with the guard removed from the processor to recover the resources.
  Without the fix, the problem may be circumvented by the following four steps:
  1) Power off the system.
  2) Use the Field Core Override function to increase the number of active processor cores in the system.  The Advanced System Management Interface (ASMI) "System Configuration -> Hardware Deconfiguration -> Field Core Override" panel shows the number of cores that are active in the system and it can be used to increase the number of active processor cores in the system.
  3) Unguard the failed processor.  Use the ASMI "System Configuration -> Hardware Deconfiguration -> Clear All Deconfiguration Errors" panel to restore the guarded processor. 
  4) IPL with the increased number of active processor cores and the unguarded processor.
  This problem does not pertain to the IBM Power System E850 (8408-44E) model.
  

System firmware changes that affect certain systems

• DISRUPTIVE:  On systems using the PowerVM firmware, a problem was fixed for an "Incomplete" state caused by initiating a resource dump with selector macros from NovaLink (vio -dump -lp 1 -fr).   The failure causes a communication process stack frame, HVHMCCMDRTRTASK, size to be exceeded with a hypervisor page fault that disrupts the NovalLink and/or HMC communications. The recovery action is to re-IPL the CEC but that will need to be done without the assistance of the management console.  For each partition that has a OS running on the system, shut down each partition from the OS.  Then from the Advanced System Management Interface (ASMI),  power off the managed system.  Alternatively, the system power button may also be used to do the power off.  If the management console Incomplete state persists after the power off, the managed system should be rebuilt from the management console.  For more information on management console recovery steps, refer to this IBM Knowledge Center link: https://www.ibm.com/support/knowledgecenter/en/POWER7/p7eav/aremanagedsystemstate_incomplete.htm.  The fix is disruptive because the size of the PowerVM hypervisor must be increased to accommodate the over-sized stack frame of the failing task.
• DEFERRED:  On systems using the PowerVM firmware, a problem was fixed for a CAPI function unavailable condition on a system with the maximum number of CAPI adapters and partitions.  Not enough bytes were allocated for CAPI for the maximum configuration case.  The problem may be circumvented by reducing the number of active partitions or CAPI adapters.   The fix is deferred because the size of the hypervisor must be increased to provide the additional CAPI space.
• DEFERRED:   On systems using PowerVM firmware, a problem was fixed for cable card capable PCI slots that fail during the IPL.  Hypervisor I/O Bus Interface UE B7006A84 is reported for each cable card capable PCI  slot that doesn't contain a PCIe3 Optical Cable Adapter for the PCIe Expansion Drawer (feature code #EJ05).  PCI slots containing a cable card will not report an error but will not be functional.  The problem can be resolved by performing an AC cycle of the system.  The trigger for the failure is the I2C devices used to detect the cable cards are not coming out of the power on reset process in the correct state due to a race condition.