Filename

Size

Checksum

8335_820.1642.20170327n_update.hpm

67109473

b3163ae821c40b10df517fff24391a4b

OP820.10
For Impact, Severity and other Firmware definitions, Please refer to the below 'Glossary of firmware terms' url: 
http://www14.software.ibm.com/webapp/set2/sas/f/power5cm/home.html#termdefs

8335_820.1642 / OP820.10

05/19/17

Impact:  New function      Severity:  NA

New features and functions

• • Support was added for HTTP(S) proxies when downloading resources during the Petitboot. For example, this support allows the user to set HTTP(S) proxies for use when loading configuration or boot files.

  Support for expanded time-out options for the BMC web gui/KVM sessions. The old time-out range (300 - 1800 seconds) has been changed as follows:

  • • Increased the maximum time-out from 1800 seconds to 1296000 seconds, i.e., basically increased the maximum time-out to 15 days.

  • •Decreased the minimum time-out from 300 seconds to 0. If the time-out value is 0, then it is considered an infinite time-out. KVM will not time-out and that in-turn will also keep the BMC web gui running too.

  Support for the Red Hat Enterprise Linux 7.3 OS as a OPAL bare-metal install. For more information on the features delivered with RHEL7.3, see the Red Hat information portal at http://access.redhat.com/documentation/en/red-hat-enterprise-linux

  Support was added to allow booting from a PCIe Non-Volatile Memory express (NVMe) adapter. The adapters affected are Feature code EC54 and EC56 NVMe flash adapters with CCIN 58CB and 58CC respectively.

  Support for a OPAL raw console to receive output from the PowerPC boot EPAPR (Embedded Power Architecture Platform Requirements) wrapper. This allows decompression failures inside the wrapper caused by data corruption to be reported to the user.

  Support for the CAPI Compression Accelerator Adapter with Feature Codes EJ1A and EJ1B and CCIN 2CF0. This CAPI FPGA (Field Programmable Gate Array) adapter acts as a co-processor for the POWER8 processor chip handling specialized, repetitive function extremely efficiently. The adapter is preloaded with a GZIP application and is intended to run as a gzip accelerator. The GZIP application maximum bandwidth is the PCIe3 interface bandwidth. Use of the EJ1A or EJ1B adapter requires one EC2A CAPI activation feature per system. This CAPI gzip feature does not run under PowerKVM but as a bare-metal install only for the following minimum Little Endian (LE) Linux distributions levels:

  • •Red Hat Enterprise Linux 7.2

  • •Ubuntu 14.04.5

  • •Ubuntu 16.04.1

  Support for the CAPI NVMe (Non-Volatile Memory express) Flash Accelerator Adapter with feature code EJ1K. This feature provides a PCIe Gen3 adapter with an FPGA and 1.92 TB of low write latency, non-volatile flash memory. The adapter physically is a half-length x8 adapter, but requires a x16 PCIe CAPI-capable Gen3 slot in the system unit. The system connects to the FPGA using the CAPI interface. The FPGA connects to the flash memory using NVMe, which is a high-performance software interface to read/write this flash memory. Use of the EJ1K adapter requires one EC2A CAPI activation feature per system. This CAPI Flash Accelerator Adapter does not run under PowerKVM but is a bare-metal install only for the following minimum Little Endian (LE) Linux distribution level: Ubuntu 16.04.1 and RedHat 7.3. The 1.92 TB of flash memory consists of two 960 GB M.2 NVMe modules mounted on the card. These modules are not removable. Each of the modules is treated as a separate LUN. The adapter card is designed for read intensive workloads with light write activity. Approximately 1,370 TB of data can be written over the life of each of the 960 GB M.2 NVMe modules. There is a "fuel gauge" function provided as part of Linux utilities enabling users to view remaining write capacity on each of the modules. If the maximum write capability is achieved, the adapter's replacement or any NVMe M.2 module's replacement is not covered under IBM warranty or maintenance. This feature only pertains to the IBM Power System S822LC (8335-GCA) model.

  Support was added for detecting and logging a SEL for power supply fan faults and turning on the system attention LED.

  Support was enhanced in the BMC LDAP configuration:

  • •Added support for allowing the hostname in any of LDAP fields that currently only allow IP addresses.

  • •Added support for BIND names greater than 64 characters.

  Support was enhanced for the configuration of the System and Audit Log settings:

  • •Log settings were enhanced to support "local Logs" and "remote Logs" at the same time.

  • •The "Server Address" field was enhanced to have a custom port number in addition to the default 514 port.

  • •The Syslog System log was enhanced to support a TCP configuration in addition to the UDP configuration that was already supported.

  Support was changed to for the BMC SMTP configuration to remove the "machine name" field since a SMTP relay server cannot be configured.

  Support was enhanced for the BMC SMTP configuration to allow the "Server Address" field to have a symbolic host name with a domain name or an IP address as was supported.

  Support was enhanced for the BMC DNS configuration to allow host names to have a "." included in the name and to allow the host names to be greater than 15 characters.

  System firmware changes that affect all systems
  
  

  A problem was fixed for error handling in complete resets for the PCI Host Bridge (PHB). During a complete reset, there can be a time-out waiting for a pending transaction, resulting in the PHB being marked as broken and the reset is not completed, leaving the adapters in an error state. With the fix, the PHB is fenced and the Linux kernel can retry the complete reset.

  A problem was fixed for a missing device discovery message and overly-verbose output messages during the boot. The output is now less verbose during the boot - only error-level messages are printed during Petitboot bootloader initialization. This means that there will be fewer messages printed as the system boots. Additionally, the Petitboot user interface is started earlier in the boot process. This means that the user will be presented with the user interface sooner, but it may still take time, potentially up to 30 seconds, for the user interface to be populated with boot options as storage and network hardware is being initialized. During this time, Petitboot will show the status message "Info: Waiting for device discovery". When Petitboot device discovery is completed, the following status message will be shown "Info: Connected to pb-discover!".

  A problem was fixed for Java error messages being displayed when logging into the BMC web gui. The Remote Console Preview window is no longer displayed on the dashboard, which was causing all the extra Java error messages.

  A problem has been fixed for the system MAC address being cleared with zeroes on an AC power cycle.

  A problem was fixed for "Preserve All Configuration" not working for the BMC web gui HPM update.

  A problem was fixed for not being able to change the IP address settings via Petitboot. Fixed a timing issue with the BCM5421 Controller. Without this fix, the timing issue would lead to BMC not getting any network even though all the network parameters/configuration are proper.

  A problem was fixed for PCIe Non-Volatile Memory express (NVMe) adapters not showing up after a power on of the system. The adapters affected are Feature code EC54 and EC56 NVMe flash adapters with CCIN 58CB and 58CC respectively. These flash adapters are not supported on the IBM Power System S822LC (8335-GTA).

  A problem was fixed for a system unrecoverable error that could occur if a CAPI adapter has an error exception signal and hangs when processing it. This is a rare problem that requires multiple error exceptions to the adapter in a short period of time, causing a deadlock in the adapter.

  A problem was fixed for CAPI adapter errors that caused a system processor to be called out and guarded instead of the CAPI adapter unit. The errors that cause this problem are the rare fatal adapter errors, so the problem should be infrequently seen. With the fix, the failing CAPI adapter is guarded after the checkstop instead of the system processor.

  A problem was fixed for looping error processing for some hardware failures where OPAL-PRD becomes unresponsive. The loop has been fixed to prevent repeated error messages and system slow-down. The error message "0xdeadbeefdeadbeef" was added so it is known when the error handling in OPAL-PRD has failed.

  An error message was changed for a problem where the SLW (Sleep Winkle) timer gets stuck and the firmware falls back to OPAL pollers. The previous error message was "Stuck with odd generation !". The new message to the SOL console is "SLW : timer stuck, falling back to OPAL pollers. You will likely have slower I2C and may have experienced increased jitter." These messages can be safely ignored at this time until a future firmware release resolves the issue of the stuck timer. The error only occurs when running test procedures that stress the hardware.

  A problem was fixed for ipmitool "mc reset cold" failing to reset the the BMC service processor. The ipmitool "mc reset cold" will stop working after the user issues power on when system is already on or after user tries to do power off when system is already powered off. The problem circumvention is to run the ipmitool "mc reset warm" command to the BMC which will restart the service processor IPMI process and clear internal flags that are preventing "mc reset cold" from working.

  A problem was fixed for slow IPMI Serial Over LAN (SOL) console connection to the server on the BMC. The problem was triggered by an incorrect handling of the definition bits for VGA and serial console output.

  A problem was fixed for the IPMI Serial Over LAN (SOL) console to the Petitboot user interface for the left and right arrow movements. When editing the command line for the kernel, the user could not go to the start of the line and then go forward one character at a time.

  A problem was fixed for the BMC integrated ethernet adapter BCM5421 connection speed being downgraded to 100 Mb/s instead of running at the expected 1000 Mb/s for the petitboot and the Linux OS. After the fix is applied, an A/C power cycle of the system is needed to activate the fix.

  A problem was fixed for a kexec-hardboot reboot of the system that caused USB devices to be lost. A system power cycle is needed to recover the USB devices when this error occurs.

  A problem was fixed for the shutdown of PCI devices that was causing spurious reboots of the system for a power off. The logical PCI devices are now removed during the shutdown.

  A problem was fixed for failures that happen when multiple Hypervisor Virtual Console (HVC) are active at the same time. On machines with more than one HVC console, any console after the first failed to register an interrupt handler since all consoles shared the same IRQ number.

  A problem was fixed for fundamental PCI resets at boot time causing the PCI adapters to not be usable in the Linux OS. No errors occur in the skiboot but the adapters are configurable once the OS is reached.

  A problem was fixed for time-out errors during the power off of PCI slots with " Timeout powering off slot ... FIRENZE-PCI: Wrong state 00000000 on slot" error message during a power off of the system.

  A problem was fixed for the system remaining in "safe" mode after an On-Chip Controller (OCC) reset. In "safe" mode, the system is running at reduced processor frequencies, affecting system performance. The OCC reset is an error recovery command that can be requested by the BMC or OPAL for certain OCC errors.

  A problem has been fixed for Fault LEDs returning to the off state after an AC cycle when a FRU had failed and was guarded. With the fix, after an AC power cycle and next power on, the Fault LED will turn on again if the FRUs are guarded.

  A problem has been fixed for a BMC cold reset that was not moving the BIOS from the golden side to the primary side.

  A problem has been fixed for systems losing performance and going into safe mode because of On-Chip Controller (OCC) timeouts collecting Analog Power Subsystem Sweep (APSS) data. This data is used by OCC to tune the processor frequency. This problem occurs more frequently on systems with large configurations that are running heavy workloads.

  A problem was fixed for a security issue on the BMC login.