Broadcom (R) NetXtreme II (TM) Ethernet Software CD for IBM eServer BladeCenter and System x NetXtreme II Gigabit and 10 Gigabit Ethernet products
RELEASE 7.10.6.0
(03/21/2014)
Support Overview ================
This is the drivers software release for all IBM eServer BladeCenter and System x Broadcom-based NetXtreme II Ethernet products, including both on-board Ethernet and adapter options.
The supported Broadcom (R) NetXtreme II (TM) chipsets include BCM5706, BCM5708, BCM5709, BCM57710, BCM57711, BCM57712, BCM57800 and BCM57810.
The supported adapter options include:
39Y6066 - NetXtreme II 1000 Express Ethernet Adapter
42C1780 - NetXtreme II 1000 Express Dual Port Ethernet Adapter
42C1790 - NetXtreme II 10 GigE Express Fiber SR Adapter
44W4479 - 2/4 Port Ethernet Expansion Card CFFh for IBM BladeCenter
44W4465 - 2/4 Port 10Gb Ethernet Expansion Card CFFh for IBM BladeCenter
All Microsoft Windows device drivers are digitally signed.
WARNING:
Any x3850M2/x3950M2 server is required to be updated to at least Version 1.05 (Build A3E145A) system BIOS, or later, prior to any device driver software update.
Caveats ======= - The following enhancement requests are not supported in this SW kit: -> CQ68018: Default to VLAN 1002 if VLAN discovery fails - FoD : FoD ready, but no FoD functions are supported at this time - Tundra : Limited to 2x10G mode operation. No 4x10G, nor 2x20G operation - VMware SRIOV support is limited to ESX 5.5 and above, and not supported on ESX 5.1 or prior. - HyperV : No Windows2012-inbox driver support in VM/Guest-OS - Storage support in T7.10 limited to 'Tundra'. - CQ70995 - Agentless Phase 1 Support "(Tundra only)" Included here on the T7.10 Gold Kit, but now shown by IBM testing to be non-functional. Will be addressed in T7.10A
Legacy ====== - E1 (57710/57711) devices are no longer supported by the 7.4 or later PXE. PXE updates on E1 devices must use 7.2.x based PXE or earlier (provided with this kit). - Windows 2003 support was discontinued as of the T7.0 program. - preventing driver install in W2K3 environment - preventing enumeration of iSCSI HBA in W2K3 environment - F6 kit is removed (no iSCSI boot over L4 path into W2K3)
Prerequisite ============ - FCoE is only supported using MFW - FCoE and iSCSI are only available in the presence of valid licenses in NVRAM - DCB/DCBX is required for FCoE. Please set NVRAM flag option #105 via Hii menu - There is no FCoE support for 10GbaseT designs
Known problems: =============== - CQ71356: Intermittent PCI-e link (GEN3) lock failure on repeated reboots - CQ71060: BACS unable to change resource reservations on NPar functions.
3rd party issues ================ - While performing shut/no shut on the switch port, it was observed that all initiators and targets were logged out from the FCoE database of FCF enabled VLAN by CN4093 switch. - After switch reboot, sometimes initiator and targets are logged out from the FCoE database of FCF enabled VLAN by CN4093 switch. Switch reboot is required to recover from this state. - In some cases, the “Application Protocol State Machine Status” for CN4093 switch port shows “Enable Operational” and “Enable Peer” status as “Disabled” for priority 3 resulting in inability to run the traffic with priority tag of “3”. This seems to be a switch issue which needs further investigation. - The “Default Port VLAN ID” of the CN4093 switch port connecting to the Broadcom FCoE initiator must be configured as “1” for switch to respond to VLAN Request. When the PVID of the switch port is configured as any value other than 1 then switch does not respond to the VLAN Request and initiator will not be able to login to the fabric.
Testing Information =================== - VMWare CMPI is NOT being tested by Broadcom PQA. - HW: -> External PHY Firmware versions for 10G Devices - 8727 : v4.06 - 8073 : v2.09 - Solarflare: v2.00 - 84833B1 : v1.52 (0x0134) - This package is provided with Storage at the committed Beta-1 level with very minimal Broadcom testing at this time - Tested with V7000 Flex storage node, DS3524, IBM XIV, DS8870 storage system and TS3310 Tape Library. Basic operation and light error injection only. - Topologies tested: IBM CN4093 in NPIV and IBM CN4093 Full Fabric mode were tested. Basic operation and light error injection only - Operating Systems tested: Windows 2012, ESX 5.1 and 5.5, RHEL6.4 and SLES11 SP3. No other operating systems tested at this time. - Compute Nodes tested: x240, x220, x440. Light testing only. - No FCoE boot support provided in this release - IBM blade BIOS/MP needs to updated to a minimum version B2BE122D/1AOO40E before installing Tundra adapter to prevent system from becoming unresponsive during boot.
Notables ======== - 10G MBA driver release no longer supports RPL protocol. All files with xMMBA.NIC and xMMBA.LOM are removed - iSCSI Initiator: [1] when using T6.x driver, there is *only* iSCSI Initiator support for the following Linux OS's: - RH6.0 and RH5.5 - SLES11-U1 [2] for any OS prior to RH5.4, or SLES11.0 the T5.2 driver set will need to be used. [3] Linux iSCSI VLAN tag dependent on L2 VLAN tag, must also be on same subnet [4] Linux iSCSI Path MTU not yet implemented [5] No routing support for SLES 11.3 ISCSI.
- The T6.2 based Solaris driver will support both 57711 and 57712. - The T6.2 based Solaris driver will not support Solaris10 U8, however it will support Solaris10 U9. - BASP w/ RSS enable can support a maximum of up to 4 processors - Broadcom does not recommend nor support enabling (manually) TCP Chimney Offload in Guest Operating Systems (GOS) running in Hyper-V on Windows Server 2008 R2 - The default Flow Control setting in the Linux driver for 5771x 10g Fiber devices is now set to disabled.
Disclaimer ========== - FCoE: Broadcom only officially supports the FCoE and DCB feature set on adapters that are licensed for FCoE, and for which EMC e-Labs certification is already complete or in progress. Broadcom will *not* provide field support for unqualified platforms. Please refer to the EMC e-Labs certification matrix for existing support matrix postings, or contact your Broadcom liaison to inquire about configurations currently in the qualification queue for eventual support. Exercising either FCoE or DCB on an unsupported platform may lead to unpredictable results.
Operating System coverage: ========================== Windows: - Server 2008 Enterprise Edition (SP2 and R2 w/ SP1) - Server 2008 Standard Edition (SP2 and R2 w/ SP1) - Server 2008 Datacenter Edition (SP2 and R2 w/ SP1) - Server 2008 for Embedded Systems (Base and R2 w/ SP1) - Server 2008 Web Server (SP2 and R2 w/ SP1) - Small Business Server 2011 (SBS V7) – Base - Small Business Server 2011 Essentials (Aurora) x64 - Server 2012 Enterprise Edition x64 (and R2) - Server 2012 Standard Edition x64 (and R2) - Server 2012 Datacenter Edition x64 (and R2)
Software Compatibility ====================== This release of software supports: - BCM5706 NetXtreme II Gigabit Ethernet Adapter - BCM5706S NetXtreme II Gigabit Ethernet Adapter - BCM5708 NetXtreme II Gigabit Ethernet Adapter - BCM5708S NetXtreme II Gigabit Ethernet Adapter - BCM5709 NetXtreme II Gigabit Ethernet Adapter - BCM5709S NetXtreme II Gigabit Ethernet Adapter - BCM57710 NetXtreme II Gigabit Ethernet Adapter - BCM57711 NetXtreme II Gigabit Ethernet Adapter - BCM57712 NetXtreme II Gigabit Ethernet Adapter - BCM57800 NetXtreme II Gigabit Ethernet Adapter - BCM57810 NetXtreme II Gigabit Ethernet Adapter - BCM57811 Netxtreme II Gigabit Ethernet Adapter - BCM57840 NetXtreme II Gigabit Ethernet Adapter
Package NOTES:
(1) When updating drivers via UXSPi, the utility will prompt version number(s) of drivers loaded on system that are pertinent to the Operating System being updated. After the utility completes and drivers are updated, there may be some driver version(s) where the New Version does not match the Installed Version. This may be because the adapter corresponding to the driver is not installed. Adapters must be installed to gain proper version numbers from the utility. Please disregard version numbers that do not pertain to the adapter(s) you have installed.
(2) When extracting files from this package, Windows 2008 R2 extract to floppy may not work. This is due to change in Windows security settings, and may become a limitation.
Version history between CD T7.10.4.1 Gold release and previous CD T7.8.4.3 RC release: ================================================================================
Software CD T7.10.4.1 Previous CD T7.8.4.3 Version Version
Key Fixes & Enhancements with this T7.10.4.1 sw CD release include: =================================================================== Highlights ========== - CQ72462 - Potential Windows bugcheck due to BASP spinlock deadlock. - CQ72483 - System hangs or continually reboots with SoleRunner using T7.8 UEFI firmware. - CQ68020 - Add support for FCoE boot over UEFI. - CQ71757 - Only mark in the consumed table entries for those items that are enabled - CQ71758 - Put WWN pushed by IFM to the actual VF where FCoE is enabled. - CQ71189 - Add new command to ediag to support external I2C NVRAM
NOTES:
(1) When updating drivers via UXSPi, the utility will prompt version number(s) of drivers loaded on system that are pertinent to the Operating System being updated. After the utility completes and drivers are updated, there may be some driver version(s) where the New Version does not match the Installed Version. This may be because the adapter corresponding to the driver is not installed. Adapters must be installed to gain proper version numbers from the utility. Please disregard version numbers that do not pertain to the adapter(s) you have installed.
(2) When extracting files from this package, Windows 2008 R2 extract to floppy may not work. This is due to change in Windows security settings, and may become a limitation.
*******Currently known issues and limitations with Broadcom NetXtreme drivers and devices. ********
1. BASP (the Broadcom Advanced Server Program) will appear to be bound to all adapters in the system once loaded. This binding is normal and is used only to determine compatibility with BASP. The binding is not active, as indicated by the gray background of the selection box. This pre-binding will not interfere with bindings to other software protocols.
2. The NetXtreme II drivers have two parts, the VBD driver and the NDIS Client driver. This combination of drivers can be used for WinPE installations as long as the WinPE image is built with the /PnP option. WinPE images that are booted with the -miniNT option will only load the NDIS driver so NetXtreme II network devices will not be functional in this environment. Use the -winpe option instead of the -minint option to boot WinPE when using NetXtreme II devices or use the special RIS driver contained in the distribution package in the NX2_RIS_Driver directory of the unpacked files.
3. Specifying network settings for NetXtreme II devices using the Microsoft unattended answer file does not work using the PCI Bus/Device/Function designation to specify adapters. This is due to the NDIS personality of the NetXtreme II device being a virtual device which means it does not have a physical Bus/Device/Function address. MAC address designation does work and may be used in conjunction with Bus/Device/Function for a complete solution. A program called netset.exe from Microsoft, available in the Windows 2000 Server Resource Kit, can also be used to apply network settings after the unattended install. Netset.exe uses the answer file to apply the network settings specified to the adapters in the system. See Microsoft Knowledge Base article 268781 for more information on netset.exe.
4. The first time diagnostics are run in a Windows environment on NetXtreme II devices, the diagnostic driver must be installed. Windows will either automatically install the driver, in which case, no action is required by the user, or Windows will ask to be directed to the INF file for the diagnostic driver. If you are prompted to install the driver, please direct the Windows dialog to search for the driver in the directory where the Broadcom driver package was unzipped. You may also search your hard drive for the file “bxdiag.inf.”
5. When loading drivers through the PNP (Plug and Play) method, users can occasionally end up loading the RIS drivers accidentally. Generally this occurs when a user is asked to point to the proper driver for the device, and the user selects the folder containing the RIS drivers as the driver for the device. This can result in network connections that fail to communicate across the network, as the NDIS driver will fail to bind to the devices. The most obvious way to identify the problem scenario is to open BACS and look at the network adapters displayed. Adapters that have the RIS drivers installed will be grayed out. Observing the driver loaded for the System Device portion of the Broadcom NetXtreme II devices will show that the RIS driver is loaded. Additionally, attempts to load the drivers using the installer will result in NDIS drivers displaying a yellow bang, as they’re not able to bind to the VBD driver. The easiest way to correct the problem is to run add/remove programs on all installed Broadcom devices, and then re-run the installer.
6. Windows Pre-installation Environment (WinPE for short) does not contain native support for 5708, 5709, 5716, 57710 and 57711 devices on operating systems prior to Windows Server 2008 R2. In order to “inject” driver support into Windows PE, refer to the steps provided by Microsoft at: http://support.microsoft.com/kb/923834/en-us. Broadcom 5708 Network Controller support is available as a native driver package in the Windows Server 2008 OS.
7. Network Teaming limitation with iSCSI offload: When iSCSI offload is enabled on Broadcom NetXtreme II NIC's or LOM's, only SLB teaming mode is supported. In order to create other teaming modes like 802.3ad and FEC/GEC, you will need to disable iSCSI offload. You can disable the iSCSI offload by a. launching the Broadcom Advanced Control Suite 3 (BACS 3). b. select the adapter you want to add to the team c. click the configurations tab, extend resource reservations. d. Uncheck iSCSI resource.
8. When iSCSI devices are installed on Windows Server 2003 or Server 2008, they will have the same friendlyname. View devices "by connections" in BACS3 on for managing iSCSI devices.
9. Restoring the team configuration with static IPv6 addresses may fail. Use either DHCP addresses for team configuration, or restore the team and re-apply the IP address manually if this issue is encountered.
10. When working in BACS while being logged into W2K3 x86 with a non-administrator account, the same icon that is used for a non-teamed adapter is being displayed for a teamed adapter. This is due to the fact the non-admin account cannot query the driver due to insufficient privileges.
11. Windows Server 2008 R2 is expected to support the ability to utilize the TCP Offload Engine (TOE) on child partitions for supported Guest operating systems. However, the stack supporting this feature was not fully matured upon the completion of this Windows drivers release. As a result, we do not recommend enabling this capability with these drivers until an update to resolve these issues are released. This is now understood to be a third-party issue.
12. After upgrading to this version of the software in W2K8 we have noticed that during a team restore in BACS3 that not all of the IPv4 addresses for VLANS are restored properly. This may require that the team be recreated and saved.
13. Remote install of Windows Server 2008 R2 over non-offload path (MSFT stack) with new Broadcom adapters, e.g. 57711, which are not supported by the inbox driver will fail. The problem is due to a conflict between the inbox (boot.wim) driver not able to support the new adapters. Although the installer provides a newer driver externally during the installation, the OS reverts back to the inbox. Please refer to the workaround provided through the following link:http://attachments.wetpaintserv.us/JA9hJcVpHrEi11mJ19vYVg%3D%3D23856.
Alternatively, Microsoft’s proposed solution is to build a custom install DVD by extracting the boot.wim and overwrite the .INF and .SYS files with Broadcom’s newer driver set. Please read Microsoft’s WAIK documentation for instruction process.
Note: Remote install of Windows Server 2008 R2 over Broadcom iSCSI Offload path does not have an issue.
14. Flow control statistics (MAC Sent X-ON, MAC Sent X-OFF and MAC Rx w/ no Pause Command) are reported incorrectly within BACS for 57710 based adapters. There is no functional impact and device is working as expected.
15. Receive Side Scaling (RSS) on TCP offloaded connections require a minimum of 8 gigabytes of RAM and only works for the 64-bit version of Windows Server 2008 and Windows Server 2008 R2. This feature is not supported in Windows Server 2003 due to a limitation in this operating system. RSS on non-offloaded connections, however, is supported for Windows 2003 SP2 and later.
16. WoL with NPar: When using Wake on LAN (WoL) with NIC Partitioning, all of the devices that comprise a port must be shut down to cause the WoL configuration to be written to the controller during operating system shutdown. If only a single function is shut down, the WoL configuration might not be written to the controller, and the controller might not wake.
17. Editing BASP teams: There is a known problem where editing a BASP team in BACS can sometimes cause a failover to occur on another team. This is currently believed to be a Microsoft issue, and Broadcom is pursuing a fix or workaround.
18. NC-SI with IOMMU: There is a known problem using NC-SI with IOMMU enabled, that can cause link down/up under stress conditions. If this occurs, disabled IOMMU in the BIOS.
brcm_dd_nic_7_10_6_0_windows_32-64_chg
Mellanox Technologies
=============================================================================== Mellanox OFED for Windows UpdateXpress Release Notes for IBM Version MLNX_WinOF 4.61 Last Modified on March 27, 2014 ===============================================================================
THIS HARDWARE, SOFTWARE OR TEST SUITE PRODUCT (PRODUCT(S)) AND ITS RELATED DOCUMENTATION ARE PROVIDED BY MELLANOX TECHNOLOGIES AS-IS WITH ALL FAULTS OF ANY KIND AND SOLELY FOR THE PURPOSE OF AIDING THE CUSTOMER IN TESTING APPLICATIONS THAT USE THE PRODUCTS IN DESIGNATED SOLUTIONS. THE CUSTOMER'S MANUFACTURING TEST ENVIRONMENT HAS NOT MET THE STANDARDS SET BY MELLANOX TECHNOLOGIES TO FULLY QUALIFY THE PRODUCTO(S) AND/OR THE SYSTEM USING IT. THEREFORE, MELLANOX TECHNOLOGIES CANNOT AND DOES NOT GUARANTEE OR WARRANT THAT THE PRODUCTS WILL OPERATE WITH THE HIGHEST QUALITY. ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT ARE DISCLAIMED. IN NO EVENT SHALL MELLANOX BE LIABLE TO CUSTOMER OR ANY THIRD PARTIES FOR ANY DIRECT, INDIRECT, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES OF ANY KIND (INCLUDING, BUT NOT LIMITED TO, PAYMENT FOR PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY FROM THE USE OF THE PRODUCT(S) AND RELATED DOCUMENTATION EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
Mellanox®, Mellanox logo, BridgeX®, ConnectX®, CORE-Direct®, InfiniBridge®, InfiniHost®, InfiniScale®, MLNX-OS®, PhyX®, SwitchX®, UFM®, Virtual Protocol Interconnect® and Voltaire® are registered trademarks of Mellanox Technologies, Ltd.
Connect-IB™, ExtendX™, FabricIT™, Mellanox Open Ethernet™, Mellanox Virtual Modular Switch™, MetroX™, MetroDX™, ScalableHPC™, Unbreakable-Link™ are trademarks of Mellanox Technologies, Ltd.
All other trademarks are property of their respective owners.
=============================================================================== Table of Contents =============================================================================== 1. Overview 2. Changes in This Release 3. Supported Firmware and Firmware Tools 4. Change Log History 4.1 WinOF VPI v4.60 Main Changes (from v4.55) 4.2 WinOF VPI v4.55 Main Changes (from v4.40) 4.3 WinOF VPI v4.40 Main Changes (from v4.3) 4.4 WinOF VPI v4.3 Main Changes (from v4.2) 4.5 WinOF VPI v4.2 Main Changes 5. Known Issues 6. Bug Fixes History 6.1 WinOF VPI v4.61 Bug Fixes 6.2 WinOF VPI v4.60 Bug Fixes 6.3 WinOF VPI v4.55 Bug Fixes 6.4 WinOF VPI v4.40 Bug Fixes 6.5 WinOF VPI v4.2 Bug Fixes 7. API Changes 7.1 WinOF VPI v4.61 API Changes: 7.2 WinOF VPI v4.60 API Changes: 7.3 WinOF VPI v4.55 API Changes: 7.4 WinOF VPI v4.40 API Changes:
=============================================================================== 1. Overview =============================================================================== This document addresses the Mellanox WinOF VPI driver package distributed for Windows Server 2012 R2 (x64 only). Mellanox WinOF VPI is composed of several software modules that contain an InfiniBand and Ethernet driver. The Mellanox WinOF VPI driver supports Infiniband and 10GB Ethernet ports. The port type is determined upon boot based on card’s capability and user setting.
=============================================================================== 2. Changes in This Release =============================================================================== WinOF VPI v4.61 provides the following changes and new features:
- Generic - Improved information in event log when a bad cable is detected - Improved resiliency on error flow in Ethernet, IPoIB and bus drivers - Performance - Added support OF IPv6 to all nd_*_* tests - Enabled optimal interrupt moderation values in SR-IOV VF mode - Stopped using NdisQueryNetBufferPhysicalCount to improve CPU utilization IPoIB - Enabled searching for IBAT routes based on dest only instead of src,dest and added a mechanism preventing memory growth in IBAT entries - Allowed any number of RSS processors, not only a power of 2 - Ensured SR-IOV mode is not enabled for IPoIB ports, which resulted in confusing message in event log Ethernet - Added support for "unknown" link state indication - Added support for DMA checks by driver verifier on SR-IOV Virtual Function - Added support for NVGRE over LBFO Team - Improved performance of handling change receive ring affinity request - In SR-IOV mode, improved resiliency to driver failures in the Virtual Machine which could result in driver load failure in VM - Improved resilience in VF to PF communication in SR-IOV mode - Improved structure of INF file for SR-IOV Physical and Virtual Functions =============================================================================== 3. Supported Firmware and Firmware Tools =============================================================================== - Firmware: - ConnectX®-3 and ConnectX®-3 EN QDR/FDR10/FDR - Firmware v2.31.2600 and above - ConnectX®-2 and ConnectX®-2 EN SDR/DDR/QDR - Firmware v2.9.1200 and above
- Tools: - MFT (Mellanox Firmware Tools) version 3.0.0 and higher
=============================================================================== 4. Change Log History =============================================================================== 4.1 WinOF VPI v4.60 Main Changes (from v4.55): ------------------------------------------------------------------------------- - Enabled configuration changes saving upon Inbox and previous releases - Enabled CIM installation as a standalone package - Added support to IPv6-to-all nd_*_* tests - Added the ability to rearm a CQ in the kernel - Added the ability to handle LID changes - Added the STATUS_CONNECTION_REFUSED return value upon connection timeout. - Added the ability to handle LID changes - Added support for iSCSI boot over IPoIB - Added DSCP support over IPv4 (Requires firmware v2.30.8000 and above) - Added traffic profile - Added ControlledBy association to IBPort - Changed connection timeout behavior - Changed the report link speed zero in case of disconnected network adapter - Modified the CQ size to prevent CQ overrun - Removed support for the following configurations: - ModeFlags - SingleMsixNum - MultiMsixNum - SingleEqNum - MultiEqNum - MaxContQuant - SlaveNum - DebugLevel - DebugFlags - UsePrio - NumFcExch - EnableQoS - BlockMcastLoopBack - InterruptFromFirstPacket - ProbeVf - Fixed the team’s MAC address uniqueness in the host of the team in Windows 2008 R2 - Performance improvements in latency - Performance improvements
4.2 WinOF VPI v4.55 Main Changes (from v4.40): ------------------------------------------------------------------------------- - Added support for Windows Server 2012 R2 Operating System - Added the ParentBusPath option to each port registry key - Added a new hardware ID for ConnectX®-3 Pro NICs - Added to the UI the *EncapsulatedPacketTaskOffload option when using ConnectX®-3 Pro NICs - Added the nd_send_bw and nd_send_lat ND benchmarking tools - Added IP-IP checksum off-load support - Added Ports TX arbitration/Bandwidth allocation per port - Added Sniffer for RoCE packets - Changed the QP numbers allocation into a round-robin manner - Removed support for RecvCompletionMethod as Interrupt - Removed the LsoV1IPv4 from the registry/UI - Removed from the bus driver configuration the ‘Non-DMA’ option - Removed the TXRingNum option from the UI - NVGRE hardware off-loads with ConnectX®-3 Pro cards only - The following ND providers, MLX4ND and MLX4ND2 are installed by default - IPoIB performance improvements - The used RoCE mode set upon driver load is printed into event log message
4.3 WinOF VPI v4.40 Main Changes (from v4.3): ------------------------------------------------------------------------------- - Added a notification in the event log in case SMB is not supported in ConnectX®-2 firmware - Added the trace tool for WPP tracing - Added Transmit Side Scaling (TSS) - Added Ethernet QoS proprietary counters, diagnostics and traffic for monitoring, using Windows’ perfmon utility - Added to the MTU size the IP header size (1500 ->1514, 9600->9614). Thus the minimum Jumbo frame size is 614. - Added active_mtu field to struct ib_port_attr_t - Added the option of vstat displaying the active_mtu of the ports - Added the option of ibv_devinfo displaying the correct MTU value after it was changed - Added the option of part_man printing the adapter name when the Port GUID is set to zero - Added the option of part_man printing the leading zeroes of port GUID - Set 512 RX buffers by default - Removed TXRingNum - Removed portsetting registry key during uninstall - Removed the QoS attributes when disabling QoS - Changed the perf_tuning setting to achieve a better performance tuning - Enabled MaxRssProcessirs support of the following values: 1, 2, 4, 8, 16, 32,64 - Interrupt moderation supports the following profiles: - Low Latency - Moderate - Aggressive in addition to old values that are not supported anymore - Made mlx4_bus and Ethernet devices removable - Network Direct: Added support for NDv2 - Network Direct: Set the default ND provide value to mlx4nd2 - Network Direct: Removed the option of NDK registration failure requiring a reboot of the machine to register it again - Network Direct: Added the option of sensing the incoming Read messages according to the device capabilities when creating an EndPoint limit - Allowed registration of a large Memory Region which is splitted to many segments - Prevented displaying a message to upgrade the firmware for OEM NICs if it has the latest firmware version
4.4 WinOF VPI v4.3 Main Changes (from v4.2): ------------------------------------------------------------------------------- - Added support for a new provider called MLX4ND, which supports both NDv1 and NDv2 interfaces - Enabled performance tuning running according to the operating systems that are running over it - The keywords added to the registry in NDIS support Windows 2012 are: - RssMaxProcNumber - NumRSSQueues - RSSProfile - The keywords added to the registry In NDIS support Windows 2008 R2 are: - RssBaseProcNumber - MaxRssProcessors - RssBaseProcGroup - The rest of the keywords are added in all versions of NDIS. This change is based on: http://msdn.microsoft.com/en-us/library/windows/hardware/ff570864(v=vs.85).aspx - RoCE MTU value is no longer set to 1024 by default - All options stay as they are and can only be chosen if they were selected explicitly in the UI/registry - The current default state is as follows: The value is now derived from the MTU (or MaxFramSize, or Jumbo Packets value) and they are all aliases for the same value). The value is aligned to 256,512,1024,2048 in a way that it will be rounded down to the nearest power of two of the ETH MTU. - Added ibdiagnet utility support
4.5 WinOF VPI v4.2 Main Changes: -------------------------------------------------------------------------------
- Added support for K-GROUPS processors (more than 64 processors support) to allow assignment of MSI-X affinity for multiple processor groups. - Added an event log to indicate driver failure upon start if there are two HCA burned with the same GUID. - Added firmware upgrade support as part of the setup process. The setup burns the new firmware only on Mellanox cards. Firmware burning failure does not prevent the driver’s installation, therefore, it will show a warning. In this case, it is recommended to update the firmware manually. - Added Registry key documentation to the setup package - Added support for BlueFlame. BlueFlame is now the default working mode for all packets that have a descriptor which fits into a BF register (currently 256 bytes). - Added support for RSS functionality on available processors numbers. Used to be restricted to start at the first processor. - Changed RSS registry defaults to give better out of the box performance - Added a performance UI to tune performance under various scenarios - Added a tool to tune performance under various scenarios - Added support for multiple TX rings - Added an option to verify that the number of multicast groups used is no higher than the firmware limits - Added support for multiple TX rings. The new driver advertises its TSS capability (using multiple TX rings). This feature will not be used when communicating with older version drivers (such as the inbox driver). This way IPv4 communication in a mixed environment that includes the inbox driver should work. IPv6 communication may be problematic in such an environment. - Added support for VM migration. When a VM is migrated from one host to another it does not send a new ARP to hosts it already knew when it was sitting on the first host. In order to prevent communication interruption to such hosts, the driver will initiate an ARP to learn them again. - Added mcast addresses according to the Windows mcast table in addition to mcast addresses from igmp\mld queries, in order to connect to such mcasts that are not reported by the queries. - Added support for the part_man tool. It can create only one additional IPoIB interface per port GUID. This feature can be utilized to use different IPoIB interfaces for SMB and Hyper-V. The virtual ports configuration is removed during uninstal - Added the following tools to the setup package: - ibv_asyncwatch.exe - ibv_atomic_bw.exe - ibv_atomic_lat.exe - ibv_devinfo.exe - ibv_rc_pingpong.exe - ibv_rdma_bw.exe - ibv_rdma_lat.exe - ibaddr.exe - ibcacheedit.exe - iblinkinfo.exe - ibqueryerrors.exe - ibsysstat.exe - saquery.exe - smpdump.exe - Modified RSS cores and changed VMQ affinity on the fly - Set an adequate number of MTTs to map all physical memory - Set QoS settings only for ConnectX-3. Changing the hardware ID, forces the OS to install new device and re-build the registry keys. - Allocated firmware and ICM memory in chunks of non-paged memory instead of using contiguous physical memory. - Allocated memory used in receive flow with the same affinity of the handling processor for faster access - Changed device names in device manager and their hardware IDs. The changes were made to distinguish between ConnectX®-2 and ConnectX®-3: - for ConnectX-2: MLX4\ConnectX-2_Eth and IBA\ConnectX-2_IPoIB - for ConnectX-3: MLX4\ConnectX-3_Eth and IBA\ConnectX-3_IPoIB - Enabled configuration of TxRingNum registry key from the UI - Improved the "Port Protocol" dialog - Improved performance in virtualization when using VMQ - Optimized code performance - Increased send parallelism - Statistics parameters are now directly read from hardware instead of being calculated by software - Use "BlueFlame" registry key to enable\disable this feature. - Removed some race conditions in IPoIB mcast handling - Removed redundant LID_CHANGE events. This fixes the redundant reconnection of all IPoIB mcasts. - Removed affinity restriction to group 0 for NDK - Installed ND by default
=============================================================================== 5. Known Issues =============================================================================== - Pinning all the physical memory (used by RDMA operations, such as register memory, pin user memory) on the machine, on Operating Systems prior to Windows Server 2012, may cause the machine to hang. - When the tracer tool traces many events, it may consume a large amount of memory (up to several GB ram). - The displayed MAC address in the DHCP server of virtual IPoIB I/F may displays a wrong data (FF00........) although the I/F is still fully functional. - When running applications that use ND or libibumad (such as OpenSM) the system might get to a unstable state when trying to shutdown/restart/hibernate it. - Activating NC-SI in WinOF v4.60 may cause driver’s loading failure when using an older firmware version than 2.30.8000. - When trying to use the following WMI classes and operations, an error or empty messages will be displayed: - MLNX_Card - MLNX_FirmwareIdentity - MLNX_FirmwareInstallationServic - MLNX_Realizes - The maximum values returned by the ib_query_ca() function (for example: max_qp, max_mr) are the upper limits of the supported resources by the device. However, it may be impossible to use these maximum values, since the actual number of any resource that can be created may be limited by the machine configuration, the amount of host memory, user permissions, and the amount of resources already in use by other users/processes. - Running Ntttcp without the “-a X” flag (X >1) in a NIC configured with 10GbE, may cause low bandwidth in TCP single stream. - Active links disappear after changing the cable connectivity from Ethernet to InfiniBand or vise versa. - On ConnectX®-2/ConnectX®-3 Ethernet adapter cards, there is a mismatch between the GUID value returned by firmware management tools and that returned by fabric/driver utilities that read the GUID via device firmware (e.g., using ibstat). Mlxburn/flint return 0xffff as GUID while the utilities return a value derived from the MAC address. For all driver/firmware/software purposes, the latter value should be used. - When working with the default NDv1 and NDv2 providers, the following error message might be displayed: 0xC0000120 NT_STATUS_CANCELLED This error does not affect any functionality and can be safely ignored. - Changing the default ND providers may cause random errors, such as: 0xC0000238 (NT_STATUS_ADDRESS_ALREADY_ASSOCIATED) on Connect() or with 0xC0000236 (NT_STATUS_CONNECTION_REFUSED) on Accept(). These errors can be safely ignored. - The UI displays the network interface’s state incorrectly even after disabling and re-enabling the "disconnected" network interfaces. - The installation process does not close any applications running in the background, and may cause a BSOD as a result of a stuck cmd. It is recommended to close all running applications prior to upgrading the driver. - InfiniBand application that was compiled with an SDK version earlier than WinOF v4.40 is not binary compatible. Recompile InfiniBand application with WinOF v4.40 and above. ND application is backward compatible and older applications over ND do not have to be recompiled. - WMI does not work due to lack of permissions.Change the execution policy. Run:Set-ExecutionPolicy AllSigned - WinOF 4.40 and above IPoIB driver is not IPv6 compatible with earlier driver versions, including Window 8 Inbox driver. If WinOF 4.50 IPoIB node receives an icmpv6 message from the legacy IPoIB node, the following event will appear in the event log: "IPoIB driver detected a possible incompatibility with Windows 8 inbox IPv6 support due to which there is no IPv6 connectivity”. - Functionality such as CQ to EQ; Auto Sensing, QoS are not supported in ConnectX®-2 adapter cards. - On rare occasions, as a result of port configuration change (IB/ETH) the UI may get stuck for up to a few minutes. This effect does not require any user action. The UI returns to its proper functionality after a few minutes. - WinOF Inbox driver does not support upgrade. When installing WinOF v4.40 and above on a Windows Server 2012 and above machine, the Inbox driver is uninstalled prior to starting the new installation and any previous configurations is lost. The Inbox driver will be reinstalled automatically when the new driver is uninstalled. - Without separate ports for each stream, WinSock multiplexes every packet to every subscriber socket and then filters it out . - A virtual IPoIB interface, created by the part_man utility, reports an Active state when the physical link is in the Initializing state and OpenSM is not running in the subnet - The "Packets Received Discarded" and "Packets Received Errors" counter may display wrong results. - Firmware upgrade may fail during installation if there was a prior firmware upgrade on the machine without a reboot after it. A firmware upgrade failure does not fail the whole installation. - The drivers’ functionality is limited up to 1TB of memory and up to 128 cores. - Connection failure on ND tests while machine A have IBAL provider and machine B have MLX4ND provider. - Hibernate and Sleep are not functional when user-space is using its resources. - Calling ib_join_mcast() with timeout_ms = 0 may result in a BSOD. - IPoIB does not support: - MAC address change - QoS (packet priority) - Load balancing and fail-over (LBFO) - Connected Mode - Partition - Memory registration on 32bit machines is limited to up to 256GB. - In an interoperability environment that has both Linux and Windows OSs, the MTU value must be the same, otherwise packets larger than the minimum will not go through. The default MTU for Linux is 2K and for Windows is 4K. - Old InfiniBand application must be recompiled with the latest SDK version. - OpenSM does not run as a service during installation since the SM requires the GUID parameter to decide which port to work on. Setting it on setup causes it to work only on the first port and not the others. - ibportstate does not work on RoCE ports - ibdiagpath may crash on Hyper-V machines - If an application which uses InfiniBand runs while the driver is being restarted, a bluescreen or an NMI may occur. - If OpenSM is up during driver restart on the same machine, it might stop working when the driver is being backed up. To resume operation, stop OpenSM and restart the driver while OpenSM is down. - When configuring one of the ports to be IPoIB, Mellanox device appears in the Eject Devices. - Sometimes during traffic, the latency of the IPoIB can get higher (this behavior is inconsistent). Set the following registry in the IPoIB network interface: RecvIntModTime=0 Please note, this action increases the CPU utilization. No communication between the physical interface and a VM that uses vSwitch created over virtual IPoIB (Pkey), and vice versa. - Disabling the "Priority & VLAN tag" in the UI which VLANID is configured, may result in sending packets with the configured VLANID. Remove the VLANID before disabling the "Priority & VLAN tag". - When working with LBFO, the teamed interface disappears after machine reboot. The issue applies to Windows Server 2012 R2 - WakeOnMagicPacket registry key is not added to the registry although WoL is supported by the driver and by the NIC. - When the ports of the device are configured as Ethernet only, ibstat/vstat may display wrong information. - High multicast drop rate on multicast storming. Use "Multicast traffic" tuning option under the performance tab. For further information, please refer to section “Tunable Performance Parameters” in the User Manual. - Driver installation requires deletion of the mlx4_bus.sys file in Windows Server 2008 R2 and WLH OSs when using the PXE package. Delete the mlx4_bus.sys file and reboot the machine to install the driver. - When there is a stress in TCP connection establishments, some of those connections may fail. Increase the Ring queue sizes: - ReceiveBuffers - controls the receive ring size - TransmitBuffers - controls the transmit ring size - Wake on Lan (WoL) cannot be disabled on NICs which supports it. - The DCB component specifies a default traffic classification that is applied to all egress packets that do not match other classification conditions. In this case, the network adapter assigns the IEEE 802.1p priority level that is associated with the default classification to these egress packets. The default traffic classification has the following attributes: - It has a traffic classification condition of type NDIS_QOS_CONDITION_DEFAULT. - It is the first traffic classification defined in the array of NDIS_QOS_CLASSIFICATION_ELEMENT structures. - RDMA Activity counters do not count during NetworkDirect RoCE traffic. - In SR-IOV mode, all ports are in pure Ethernet mode and RDMA is not supported on either port. - When NVGRE off-load is enabled, the GRE traffic cannot be accepted as a regular L2 traffic and requires special L2_TUNNELING steering rules. In such case the GRE packets are dropped or directed to promiscuous queue. - GRE traffic steering by inner MAC and by outer MAC simultaneously is currently not supported. Configure steering or by inner MAC, or by outer MAC. - If VMQ set filter requests are accepted without a GRE flag (i.e. requested steering by outer MAC), the GRE packets do not reach that VMQ. Set the bus driver registry keyAcceptGREbyOuterMAC_P1/2 per port to accept GRE traffic by outer MAC and to duplicate L2 steering rule to L2_TUNNELING rule for each VMQ set filter request without GRE flag. Note: For regular NVGRE Hyper-V scenarios the value of the registry key below must be set to 0:AcceptGREbyOuterMAC_P1/2 - Running Quality of Service (QoS) commands without the parameter "-PolicyStore ActiveStore” may cause machines to load without Quality of Service policy. Store the QoS policy in the ActiveStore - RoCE does not support: - Traffic cannot go through the router. It works in the same subnet only - Multicast traffic - VLAN - Layer 3 feature - Using different versions of RoCE in your cluster is not supported. Use the same RoCE version in all the cluster in the Ethernet ports. - When using WinOF 4.40 or above, low throughput will be seen on 40GbE adapters when QoS is enabled. To solve it Disable QoS when it is not in use: - Open a PowerShell prompt. Run: - Disable-NetAdapter-Qos -name where is - e.g. "Ethernet 1" - Perf_tuning is supported only when one of the two NUMA nodes are in use. - Execution of nd_write_lat over mlx4ndv1 or mlx4ndv2 may never end. Press CTRL-C and kill the test. - Running performance benchmarks for a short period of time (< 1 sec) may provide bad latency in IPoIB and Ethernet. Set “Rx Interrupt Moderation Profile” and “Tx Interrupt Moderation Profile”, to “Low Latency” to avoid bad latency. Note: This may increase CPU utilization. - When the vSwitch is detached from the ETH\IPoIB device while the driver is disabled, the device does not reacquire the static IP it had before the attachment of the vSwitch. When the vSwitch is attached to the ETH\IPoIB device while there is no link, it will not receive the device IP when the link is back up. - After attaching the vSwitch to the ETH\IPoIB device, changing the “Jumbo Packet” registry key on the ETH\IPoIB device does not affect the vSwitch configuration and vice versa. For example, if the user sets the “Jumbo Packet” on the ETH\IPoIB device to X, and the “Jumbo Packet” on the vSwitch to X+Y, X+Y sized packets will be passed from NDIS down to the driver and they will be dropped by it. Reattach the vSwitch to sync with the value set in the ETH\IPoIB device. - Unexpected behavior might occur when running in a virtualized environment and creating two virtual switches bound to each of the ports of a dual port NIC and then using both of them with two vNICs from the same VM. - When moving an IPoIB interface in a VM from non-VMQ to VMQ or from VMQ to non-VMQ, a reset to the NIC may occur and in the event log the following message will appear: ‘device reports a "CQE error" on cqn qpn Status . Therefore, the HCA Nic will be reset. (The issue is reported in Function ). For more information refer to details.’ - In IPoIB when using long Multicast traffic from a Virtual Machine (VM) to an external host there might be up to 0.5% loss in 5% bursts - Hyper-V is at low bandwidth on LBFO vSwitch, Windows 2012 - In Ethernet to achieve better iperf TCP performance between a Linux VM and a Windows VM on different hosts, when using MS MUX over the Ethernet driver, use the non VMQ mode for the VMs. - After disabling and enabling a port on a guest, a ping to it may be renewed after a minute. The ARP requests sent by Windows are less frequent as the time passes. If the guest port was down for a while, it could take time until Windows decides to send another ARP request to it. - When VMQ is enabled after reset, the driver loads all the VMQs that existed before the reset. However, it is not guaranteed that each VMQ will receive the same QP number it had before the reset. This can cause some delay as a result of resetting before connectivity is reestablished. The delay is caused by the time it takes for the ARP table to update after initiating the Gratuitous ARP. - The IPoIB non-VMQ mode is supported only when the VMQ is enabled according to the registry values. To use the non-VMQ mode for a VM, change its settings as follow: - Press "Settings" on the VM - Go to Network Adapter -> Hardware Acceleration - Un-check the "Enable virtual machine queue" - Upgrading the driver while the UI is opened with the "ConnectX NIC device" may cause the installation process to never end. Close the UI before driver upgrade. - Rebooting the machine reboot while uninstalling WinOF, may result in installation failure. Delete Mellanox components from HKEY_LOCAL_MACHINE\SOFTWARE\Microsoft\Windows\CurrentVersion\DIFxApp\Components. The Mellanox components are mlx4eth63, ipoib6x and mlx4_bus - Canceling the installation process may leave the bus driver in a disable state. The driver appears in a yellow bang containing the following error message: “Windows cannot start this hardware device because its configuration information (in the registry) is incomplete or damaged. (Code 19)”. Scan for new hardware and reboot the machine. - Downgrade is not supported. Uninstall the current version and install the older one. Installation is stuck when Remote Desktop Session Host Windows Installer RDS Compatibility is enabled. - Occasionally, upon driver upgrade due to mismatch of driver versions, the following error message is displayed: “There is an interface mismatch between ETH driver and the bus driver. The ETH driver interface version is 50 while the bus driver interface version is 17. As a result the ETH driver has failed to start. This happened due to setup failures or partial update of the drivers. In order to resolve the issue, please reboot the computer” The message can be safely ignored. - Configuration is not restored when replacing a ConnectX-2/ConnectX-3 NIC with a ConnectX-3 Pro NIC located on the same PCI slot Clean the old network adapter configuration prior to upgrade. - Configuration can be restored only in Windows Server 2012 and above - IPv6 configuration restore is not supported - Upon upgrade, the following Registry Key values will be overwritten with the following: - ReceiveBuffers = 512 - MaxRssProcessors = 8 - RssBaseProcNumber = 0 - NumRSSQueues = 8 - RssMaxProcNumber = 63 - RssProfile = 1 - DefaultRecvRingProcessor = -1 - TxInterruptProcessor = -1 - TxForwardingProcessor = -1 - RxIntModerationProfile = 1 - TxIntModerationProfile = 1 - RecvCompletionMethod = 1 - SingleStream = 0 - TxRingNum = 8 - Upon upgrade the following Ethernet Registry Keys will be deleted: - SendCompletionMethod - UseRSSForRawIP - UseRSSForUDP - Upon upgrade the SendCompletionMethod IPoIB Registry Key value will be modified as follow: - SendCompletionMethod = 0 Upon upgrade the following IPoIB Registry Keys will be deleted: - UseRSSForRawIP - UseRSSForUDP - The empty directory %ProgramFiles%\Mellanox\MLNX_VPI\Endure DLLs is being created during the driver installation. - Uninstalling the driver on Windows Server 2008 R2 with LBFO configuration results in the appearance of a pop-up window requesting to close several running applications. Choose "Do not close applications”. This action allows the uninstallation of the driver. A Reboot may be required. Rebooting the server before uninstalling the driver when LBFO is configured will eliminate this pop-up completely. - Upgrading the driver while perfmon is open and monitoring the "ConnectX NIC device" may end up with adapters having yellow bang with code 10. Close the perfm`on before driver upgrade. - ibdump may encounter packet drops upon a burst of more than 4096 (or 2^max-burst) packets. - Packets loss is not reported by ibdump. - Running ibdump on a RoCE Ethernet port may decrease the functional bandwidth due to the overhead of creating extra copy for each packet. This may lead to packet drops on the link. Verify Ethernet flow control is enable to ensure a lossless link - Sniffing over IB ports is currently not supported
=============================================================================== 6. Bug Fixes History ===============================================================================
6.1 WinOF VPI v4.60 Bug Fixes: ------------------------------------------------------------------------------- - Ensured that uninstallation of Mellanox package in Virtual Machine leaves the system clean - Fixed an issue preventing JumboPackets registry key to be restored correctly - Fixed an issue which caused Mellanox devices to be listed in "Devices and Printers" and had "Safe Removal" UI - Fixed error statistics collection which could cause false error report - Fixed a connectivity problem between Hyper-V VMs on the same host - Fixed loopback issues in the virtualization environment - Fixed stability issues - Fixed an issue that prevented receiving ARP traffic in NVGRE mode
6.2 WinOF VPI v4.60 Bug Fixes: ------------------------------------------------------------------------------- - Fixed using CQ after VMQ is closed - Fixed bad completion of VMQ QP that was caused by malformed WR - Fixed an issue occurred when uninstalling and reinstalling the driver. The ConnectX-3 Pro Ethernet device was displayed in the Device Manager with a yellow bang (!). - Fixed an issues enabling the package’s execution in modify mode resulting in driver being disabled - Fixed an issue occurred when uninstalling and reinstalling the driver. The ConnectX-3 Pro Ethernet device was displayed in the Device Manager with a yellow bang (!). - Fixed random parsing failures of string registry entries - Fixed compilation failure of "Hello_world" in the SDK - Fixed the return value of ib_query_ca() if failed to allocate resources for operation - Fixed RoCE mode parsing - Fixed CPU utilization report in nd_*_* tests - Fixed correct bandwidth peak results in ibv_send_bw with UD QP - Fixed sync problems of bidirectional mode in ibv_read_bw/ibv_write_bw - Fixed missing completions when working with Completion Queue with single entry - Fixed unexpected behavior upon QP asynchronous event - Fixed bad completions of VMQ and NonVMQ modes in IPoIB - Fixed a failure occurred when setting the IPoIB adapter value to "SA Query Timeout" - Fixed propagation of the physical link disconnection to virtual (part_man) interface - Fixed un-initialized memory access over the VMQ - Fixed port channel teaming with CISCO switch and Fabric Extenders traffic loose in Windows Server 2008 R2 - Fixed an issue related to packets sent with corrupted VLAN header when they were meant to be untagged - Fixed unexpected behavior upon QP asynchronous event - Stability fixes - Fixed ConformsToProfiles association for SoftwareIdentity and DriverIdentity - Fixed execution of all tests which were running when executing Diagnostic tests on one instance - Fixed a failure occurred when running MLNX_Card - Fixed the printing of diagnostics log - Fixed an issue preventing from get-event to show information after disabling the PCI device
6.2 WinOF VPI v4.55 Bug Fixes: ------------------------------------------------------------------------------- - Fixed nd_*_bw to achieve better performance (memory buffer alignment) and consistent results - Fixed the issue preventing messages to be sent in VLAN 0 when using many VMQ rings - Fixed a part_man issue related to wrong statistics over virtual partman interfaces
6.4 WinOF VPI v4.40 Bug Fixes: ------------------------------------------------------------------------------- - Removed port setting registry key during uninstallation - Fixed an issue causing the setup to fail upon perf_tune failure during the installation. An error message will be printed in the installation log upon perf_tune failure. - Fixed and issue with the Mellanox adapter being shown on the USB removal menu, which caused the removal of the Mellanox adapter once removing the USB. - Fixed the perf_tuning indication of the last chosen tuning scenarios - Fixed a crash in the ib_send_lat/bw utilities caused when the port link was down - Fixed the “Restore to defaults” option in the perf_tuning tool. Now the default values are being restored - Fixed WoL support on NIC with a single port - Fixed the default RoCE configuration on NICs with a single ports - Fixed the values for the MTU and rate of the CM-REQ - Fixed miniport reset on sending scenarios - Fixed a bluescreen issue that occurred when disabling the interface after a TX stress over the VMQ - Fixed a failure of MPI/ND over InfiniBand - Network Direct: Fixed a crash occurred when more than 4 SGEs elements were used in an ND write operation - Network Direct: Fixed the swap of InboundReadLimit and OutboundReadLimit when creating an EndPoint and in Connector::GetConnectionData - Network Direct: Fixed disallowing creation of EndPoint with zero attributes in the Receive Queue - Network Direct: Fixed a failure when creating an EndPoint with zero attributes in the Receive Queue
6.5 WinOF VPI v4.2 Bug Fixes: ------------------------------------------------------------------------------- - Fixed restart issue when there are not enough MSI-X vectors for each machine core - Fixed RSS indirection table mapping building when there are less RX rings than RSS cores - Fixed a bug, preventing standard work with BAR value more than 4GB. - Fixed memory leaks - Fixed error flows causing a Bluescreen in driver startup/unload - Fixed a Bluescreen occurrence upon shutdown due to leak in active resources - Fixed a bug that prevented IBAL applications from working on machines with 2 cards - Fixed a bug that caused packet drop on remote node when the first sent packet is not an ARP or a multicast packet - Fixed IPoIB VMQ affinity update bugs - Fixed IPoIB VMQ parent queue management race during reset - Fixed a bug in passing DHCP packets from a Linux VM - Fixed bug in RDMA statistics for NDK
=============================================================================== 7. API Changes ===============================================================================
7.1 WinOF VPI v4.60 API Changes: ------------------------------------------------------------------------------- - Added a new macro for IB_MOD_QP_CHANGE_COUNTER_INDEX - Added the field state.rtr.counter_index for struct ib_qp_mod_t
7.2 WinOF VPI v4.55 API Changes: ------------------------------------------------------------------------------- - Added enumerated values for RDMA_TRANSPORT_RRDMAOE_1_5 - Added enumerated values for RDMA_TRANSPORT_RRDMAOE_2_0 - Added enumerated values for RDMA_TRANSPORT_RRDMAOE It is an alias to RDMA_TRANSPORT_RRDMAOE_1_5 - Added new functions for is_rroce(), is_xroce() - Added a new macro for IB_AC_SNIFFER - Added the field state.init.flags for struct ib_qp_mod_t
7.3 WinOF VPI v4.40 API Changes: ------------------------------------------------------------------------------- - Increased the value of VERBS_MINOR_VER - 0009 -> 000a - Added enumerated values for enum eth_link_speeds - struct ib_port_attr_t: - The mtu field was separated into two fields: - max_mtu (maximum MTU supported by the port) - active_mtu (actual MTU which the port is configured with) - Added the eth_link_speed field - Added enumerated values for WR_SEND_INV - The type of invalidate_rkey for struct ib_send_wr_t was changed from net32_t -> ib_net32_t - Added the send Write flag for IB_SEND_OPT_SKIP_DOORBELL mlnx_fw_nic_4_61_50000p4_win2k12r2_32-64_chg
=============================================================================== Emulex 16GB FC Firmware Package ===============================================================================
Firmware Version: 10.2.261.36 Supported On: IBM System x and Flex
Problems Fixed: - Server will halt on black screen during rolling reboot of Windows 2012 x64 uEFI SANboot - Server intermittently fails to boot during rolling SANboot (any OS) - Fixed a boot hang associated with server BIOS change
Incremental Interoperability: - Inventory support for the Flex xHMC (Chassis Hardware Management Console)
Known Issues: - To disable BFS after BOFM or IFM have been enabled, it must be done by setting BFS manually in the EFIBoot config utility or sending a BOFM default command to the port. This is working as expected and we are documenting it here.(159504)
Firmware Version: 1.1.43.21 Supported On: IBM System x
Incremental Interoperability: - None
Problems Fixed: - Server will halt on black screen during UEFI rolling reboot reboot cycle test with 16Gb HBA under Windows 2012 x64 - Allows installing and booting from an MSA2040 array configured in PT-PT direct mode at 16G speed. We're now able to enable Autoscan and install/boot from multiple OSes - Added support to bypass configuring pnp Boot Table in a multipath situation. Fixes Microsoft Windows Server 2008 bug. - Fixed the "Read Rev Failure" issue during POST to accept "0" address value as a valid. - Fixed the wrong VPD HBA model number display on the banner and properties, and change the proper VPD data field size. - Fixed the timeout issue on Flogi when try to select a HBA device without connecting to a target or FC switch. - System will halt during reboot cycle test under Windows 2012 x64 UEFI - Fixing PCI Function Number decoding issue in BOFM code - Boot fails during SLES11 SP3 rolling reboot - Enable BFS and update the NVRAM if BOFM target is available - Add Diagnostic support for UEFI 16Gb FC - Reconnect menu item causes a System BIOS exit request only on AMI core system BIOS - Adding port's protocol name and WWPN in browser help menu
Firmware Version: 1.1.43.12 Supported On: IBM System x and Flex
Incremental Interoperability: - IBM Flex System FC5052 2-port 16Gb FC Adapter - IBM Flex System FC5054 4-port 16Gb FC Adapter - FC support for support for ASU/Config Patterns for Flex and ASU for IBM System x & BladeCenter - Direct PT - PT connections - Extended SCSI Pass Thru Protocol - IBM Tape unit interoperability - see SSIC for more information - SLES 11.3 on IBM System x
Problems Fixed: - Windows 2012 OS will not resume from Sleep mode - Boot Device entries can be lost after exiting from BIOS setup
Firmware Version: 1.1.21.9 Supported On: IBM System x
Incremental Interoperability: - IBM Server machine types. Refer to IBM ServerProven for specifics
Problems Fixed: - Fixed FCP_CONF handling bug - No SRR sent after drop write data frame, I/O fails - Adapter does not invoke recovery after dropped read data frame - I/O throughput performance improvements - Link failover when connected to FC Tape - Cosmetic: Cleaned up miscellaneous error handling messages in debug log files - If a new device comes online and attempts to acquire an AL_PA after the link has already come up, the FC link appears to hang until the cable is pulled - Cleaned up link negotiation at 4GB/8GB/16GB speeds with specific switches
=============================================================================== Emulex FC Device Driver for Windows ===============================================================================
Device Driver Version: 10.2.261.4 Supported On: IBM System x, BladeCenter, and Flex
Problems Fixed: - ABTS sent by tape device ignored by HBA - BSOD issue during CPU Hot replace test
Device Driver Version: 2.74.016.001 Supported On: IBM System x, BladeCenter, and Flex
Problems Fixed: - Cleaned up a variety of corner case issues to avoid BSODs - Fixed IOCTL encounter bug-chk on BE3 during NPIV tests - Fixed reporting incorrect error code when creating vport on a switch without NPIV supported - Updated the default model name for 16G HBAs
Device Driver version: 2.74.009.001 Supported On: IBM System x, BladeCenter, and Flex
Problems Fixed: - Clean up non-critical legacy FC and uCNA incompatibilities - LUN masking doesn't persist after reboot - Emulex GUI crashes when closing a running Upgrade utility and re-open it in 802.3 configuration with IP or MAC - Several bugs related to creation or deletion of vports
Incremental Interoperability: - Windows 2012 - Improved debug capabilities - Tighter integration with MS Windows Design Kit
Device Driver Version: 2.70.018 Supported On: IBM System x, BladeCenter, and Flex
Problems Fixed: - Fixed sending ABTS continuously to Flogi issue - Fixed reset and link bounce issues for SLI4 FC public loop - Added support for ABTS timeout handling for both SLI3 and SLI4 - Fixed loopback test incorrectly dropping the loopback frame in SLI4 - Fixed exchange management issue in SLI4 CT receive path causing array overflow - Fixed firmware version update issue following download in link up state - Fixed problem in init_vfi completion handler that stalled fabric login on BE FCoE FW - Fixed problem that caused OCM ELS Echo Test to fail - Fixed the loopback test issue in recovering from error condition - Add persistent link down support for SLI4 FC - Fix DFC_IssueDumpMBox to manage request type 3 which includes link select option - Modified reinitialization sequence following reset to fix discovery failure post type 1 or type 3 reset - Fixed new firmware dump issue
Fibre Channel Adapter STOR Miniport Driver for Windows Release Notes February 18, 2014
QLogic Driver Release Notes
QLogic Corporation All rights reserved Package Version 9.1.11.26.b Driver Version 9.1.11.26
Table of Contents
1. Change History 2. Known Issues 3. Notices 4. Contacting Support
1. Change History
The following changes have been made to the Fibre Channel Adapter STOR miniport driver for version 9.1.11.26.
The following changes have been made to the Fibre Channel Adapter STOR miniport driver between versions 9.1.11.25 and 9.1.11.26.
* API version 1.28.0.112 (QLSDM) * [ER115248]: Read NVRAM fails after write Resolution: Added delay between write and read in api Scope : 8200 Series Adapters * [ER115315]: Unable to update EDC Resolution: Fixed wrong lengh in built-in FLT Scope : 2500 Series Adapters * [ER114713]: Poor read write performance at certain large block sizes Resolution: Fixed delay in handling SRB queue Scope : 2600 Series Adapters * Firmware version: 7.01.00 (2400/2500 Series 4Gb and 8Gb Fibre Channel Adapters) * [ER114741]: Login to tape device times out Scope : All FC and FCOE Adapter Resolution: Retry Relogin once * [ER113510]: Maximum number of GNNFT entries is limited to 128 Scope : All Adapters Resolution: Increased maximum number of GNNFT entries * [ER113482]: Failover on EMC storage not working when zoning is deactivated on switch Resolution: Fixed for not relogin after timeout * [ER112112]: SnapDrive cannot see luns through the QLE10K Scope : 10000 series adapter Resolution: Change the number of ports in AdapterAttributes to 1 * [ER112725]: Issue with processing certain response entry during stop HBA operation Resolution: Added check for HBA stop state before response queue processing * [ER112726]: Multi-queue debug trace feature Scope : 2500 and 2600 Series Adapters Resolution: Added per-queue debug information trace * Added support for the following devices (supported platforms): - PCI\VEN_1077&DEV_2532&SUBSYS_02841077 (x86, x64)
Version 9.1.11.24
* Firmware version: 5.09.00 (2400/2500 Series 4Gb and 8Gb Fibre Channel Adapters) * API version 1.28.0.105 (QLSDM) * Fix MailboxPassthru by removing restriction on writing mbx8 register [ER104361] * Added support for Skip 4G FC [ER104575] * Allow application modify "Execution Throttle" field in nvram for 2600 series adapters [ER103297] * Add Service Pack number to OSName [ER104204] * Added api to read 2600 series adapters ASIC temperature [ER102430] * Added mailbox passthru api [ER102988] * Added reading ASIC temperature for Ninja and Sensei [ER103001] * Removed Advanced tab from Device Manager properties * Added support support for full 64 bit physical addressing [ER101742] * Added support for QLE10000 series adapters
Version 9.1.10.28
* Firmware version: 5.07.02 (2400/2500 Series 4Gb and 8Gb Fibre Channel Adapters) * API version 1.28.0.101 (QLSDM) * Removed support for 2Gb speed in port supported speed for 16Gb FC in 2600 series adapters [ER100053] * Added get/set I2V registers and get/set serdes registers - OEM Adapter [ER99727] * Added support to address OEM adapter Re-design I2C access issue [ER99562] * Added adapter disable state check on timer routine re-init state handling [ER96662] * External loopback support OEM adapter when in loop mode [ER98834] * Fixed skipping OEM adapter vpd update [ER100458] * Fixed Ext Loopback test fails with Loopback connector (8Gb loop mode) [ER99702] * Fixed crash in fwdump routine [ER101039] * Limit support to node-interleaving configurations
Version 9.1.10.27
* Fixed event Viewer Error 154 (Disk I/O error) occurs when running Diskercise on FC LUN [ER97809]
Version 9.1.10.26 * API version 1.28.0.97 (QLSDM) * Added series adapters temperature handling [ER94159] * Removed driver family version support [ER95755] * Removed flash FLT error logging for 24xx/25xx [ER96144] * Returned NOT_SUPPORTED for SDForceFcMctpDump in api [ER96352] * Fixed memory access error [part of ER96188] * Fixed extra 4K in variable transfer sizes (64K, 128K) [ER94081] * Added support to relogin if the device logs out [ER93867] * Removed event logging for fw dump event [ER98897]
Version 9.1.10.25
* Added Windows Server 2012 support [ER92249] * Not to update i2c vpd if Sensei in api [ER93800] * Fixed Flash Layout Table memory usage [ER93668] * API version 1.28.0.96 (QLSDM) * Code cleanup [ER92250] * Added support for Set Driver version for OCBB and family driver version [ER92243] * Issue with LB connector attached after it was in point-to-point mode, driver was still issuing echo instead of loopback command [ER93050] * Fixed Adapter Port type in P2P mode [ER92357] * Process completed PRLI [ER89053] * Fixed multiple request single response queue initialization issue [ER95259]
Version 9.1.9.48
* Firmware version: 5.06.05 (2400/2500 Series 4Gb and 8Gb Fibre Channel Adapters)
Version 9.1.9.47
* Firmware version: 5.06.04 (2400/2500 Series 4Gb and 8Gb Fibre Channel Adapters)
Version 9.1.9.46
* Firmware version: 5.06.03 (2400/2500 Series 4Gb and 8Gb Fibre Channel Adapters) * API version 1.28.0.92 (QLSDM) * LUN disappearance after tape reboot [ER84117] * Failed to reissue some i/o's after controller failover (8Gb) [ER86660]
Version 9.1.9.29
* Firmware version: 5.06.02 (2400/2500 Series 4Gb and 8Gb Fibre Channel Adapters)
Version 9.1.9.28
* Firmware version: 5.05.05 (2400/2500 Series 4Gb and 8Gb Fibre Channel Adapters) * API version 1.28.0.87 (QLSDM)
Version 9.1.9.27
* Updated firmware version: 5.04.04 (2400/2500 Series 4Gb and 8Gb Fibre Channel Adapters) * API version 1.28.0.85 (QLSDM) * Fixed an issue where drives were not detected in direct connect topology [ER81138]
Version 9.1.9.25
* No fixes in this driver release.
Version 9.1.8.25
* Closed WMI mapping interoperability with HBAAPI.DLL issue.
Version 9.1.8.19
* Driver restores 4G Fibre Channel PCIe Device Control and Link Control registers based on saved value. * In multi-processor systems, the clustering driver timer popped as a result of unrecovered I/O through a LUN reset. [ER0000000067688 closed]
Version 9.1.8.17
* Firmware versions: 5.01.02 (4Gb and 8Gb) * API version 1.28.0.73 (QLSDM) * Driver restores 4G Fibre Channel PCIe Device Control and Link Control registers based on saved value.
Version 9.1.7.18
* Firmware versions: 3.03.25 (2Gb), 4.04.00 (4Gb and 8Gb). * API version 1.28.0.63 (QLSDM).
Version 9.1.7.17
* Added support for the following devices (supported platforms) (2.3m): - PCI\VEN_1077&DEV_8432&SUBSYS_010E1077 (x86, x64) - PCI\VEN_1077&DEV_2432&SUBSYS_014B1077 (x86, x64) - PCI\VEN_1077&DEV_2532&SUBSYS_01701077 (x86, ia64, x64) - PCI\VEN_1077&DEV_2532&SUBSYS_01711077 (x86, ia64, x64) - PCI\VEN_1077&DEV_2532&SUBSYS_01721077 (x86, ia64, x64) - PCI\VEN_1077&DEV_2532&SUBSYS_01731077 (x86, ia64, x64) - PCI\VEN_1077&DEV_2532&SUBSYS_01741077 (x86, ia64, x64) - PCI\VEN_1077&DEV_2532&SUBSYS_01751077 (x86, ia64, x64) - PCI\VEN_1077&DEV_2432&SUBSYS_004A1137 (x86, ia64, x64) * Removed support for the following devices (supported platforms): - PCI\VEN_1077&DEV_2432&SUBSYS_011C1077 (x86, x64) NOTE: Microsoft requires x86 and x64 support for all Fibre Channel Adapters. * Firmware versions: 3.03.25 (2Gb), 4.03.02 (4Gb). * API version 1.28.0.63 (QLSDM). * Added support for 64-bit QLSDM DLLs. * Enabled FDMI by default. * Added support for Atmel flash AT26DF081A. * Removed support for obsolete qd parameter. * Changed the driver to not log events caused by I/O completion errors.
Version 9.1.7.16 QL
* Added support for the Windows Server 2008 operating system. * Applied a fix to handle logout in the loop environment.
Version 9.1.7.15 QL
* Added support for the following devices (supported platforms): - PCI\VEN_1077&DEV_2432&SUBSYS_01081077 (x86, IA64, x64) - PCI\VEN_1077&DEV_2532&SUBSYS_015C1077 (x86, IA64, x64) - PCI\VEN_1077&DEV_2532&SUBSYS_015D1077 (x86, IA64, x64) - PCI\VEN_1077&DEV_2532&SUBSYS_015E1077 (x86, IA64, x64) - PCI\VEN_1077&DEV_2532&SUBSYS_015F1077 (x86, IA64, x64) - PCI\VEN_1077&DEV_2532&SUBSYS_01601077 (x86, IA64, x64) - PCI\VEN_1077&DEV_2532&SUBSYS_01611077 (x86, IA64, x64) - PCI\VEN_1077&DEV_2532&SUBSYS_01621077 (x86, IA64, x64) - PCI\VEN_1077&DEV_2532&SUBSYS_01631077 (x86, IA64, x64) - PCI\VEN_1077&DEV_2532&SUBSYS_01641077 (x86, IA64, x64) - PCI\VEN_1077&DEV_2532&SUBSYS_3261103C (x86, IA64, x64) - PCI\VEN_1077&DEV_2532&SUBSYS_3262103C (x86, IA64, x64) - PCI\VEN_1077&DEV_2532&SUBSYS_3263103C (x86, IA64, x64) - PCI\VEN_1077&DEV_2532&SUBSYS_3264103C (x86, IA64, x64) NOTE: Microsoft requires x86 and x64 support for all Fibre Channel Adapters. * Firmware versions: 3.03.25 (2Gb), 4.02.00 (4Gb). * API version: 1.28.0.50 (QLSDM). * Unique assignment of management server N_Port handle values to be for each virtual port. * NPIV (N_Port ID virtualization). * Target link speed (iiDMA). * PcieLinkCap/Stat to HbaDeviceChipProperty. * Flash layout regions for 4Gb and returned flash layout table from flash if it is valid. * Three Alternate Boot Devices and OOFR (out-of-order frame receive). * Removed QL2XHAI2: IHVs are required to use Host Bus Adapter API provided by Microsoft. * Preserved BIOS defaults when updating multi-boot image whose BIOS locations can be anywhere in the image. * Fixed checksum calculation in QLSDM when two-bios image is flashed. * Initiator discovery problem caused a large burst of RSCN activity when the NetApp does its failover. A failed GPNID command occurring during ports discovery must not cause the termination of the discovery process. [ER50975 fixed] * Handled the case where the GNNFT command was failing due to incorrectly calculated the maximum data response length. * Setting the alternate WWN and enabling it in the NVRAM did not propagate through to the Host Bus Adapter. Reading the NVRAM and reinitializing the Host Bus Adapter will handle this case.
Version 9.1.4.16 QL 3.03.21 4.00.26
* Adhered to Subsystem Id document v2.2m+. * Firmware versions: 3.03.21 (2Gb), 4.00.26 (4Gb). * API versions: 1.28.0.37 (QLSDM), 2.0.0.14 (QL2XHAI2). * Initiator discovery problem caused a large burst of RSCN activity when the NetApp does its failover. A failed GPNID command occurring during ports discovery must not cause the termination of the discovery process. [ER50975]
Version 9.1.4.15 QL 3.03.21 4.00.26
* Firmware versions: 3.03.21 (2Gb), 4.00.26 (4Gb). * API versions: 1.28.0.37 (QLSDM), 2.0.0.14 (QL2XHAI2). * Adhered to Subsystem Id document v2.2m+. NOTE: Microsoft requires both x86 and x64 support all Host Bus Adapters. * Added support for the following devices (supported platforms): - PCI\VEN_1077&DEV_2322&SUBSYS_01151077 (x86, IA64, x64) - PCI\VEN_1077&DEV_2322&SUBSYS_01161077 (x86, IA64, x64) - PCI\VEN_1077&DEV_2422&SUBSYS_1336103C (x86, IA64, x64) - PCI\VEN_1077&DEV_2432&SUBSYS_015B1077 (x86, IA64, x64) - PCI\VEN_1077&DEV_5432&SUBSYS_01551077 (x86, x64) - PCI\VEN_1077&DEV_5432&SUBSYS_01561077 (x86, x64) - PCI\VEN_1077&DEV_2422&SUBSYS_014A1077 (IA64) - PCI\VEN_1077&DEV_2432&SUBSYS_014B1077 (IA64) * Removed support for the following devices: - PCI\VEN_1077&DEV_6312&SUBSYS_01301077 (x86, x64) - PCI\VEN_1077&DEV_2422&SUBSYS_01471077 (IA64) - PCI\VEN_1077&DEV_2432&SUBSYS_01451077 (IA64) * Conformed QLSDM and the driver to the revised for "Get/Set Target Link Speed". * Increased loopback test buffer size to 2048 bytes in QLSDM. * Loopback test always failed when the NVRAM frame size is set to 512 or 1024 (Windows). Pass in transfer size instead of hard-coded value in QLSDM. [ER50975 fixed] * Handled the case where the device becomes not ready in the StartIO routine. * Fixed in SANsurfer 5.0.0 Build 4 – Driver 9.1.3.10b2—Windows IA64 SP1—HP. The Host Bus Adapter model numbers do not show up on Information page. [ER48374] * Fixed Long Erase (0x019) issues with DLT S4 and QLE2462. Error occurs (most likely a bus reset) when running a long erase that takes longer than 110 minutes on the Quantum DLT-S4 (firmware:0808) drive, which exceeds the built in 4Gb firmware timeout limit. [ER48117 (AR 48907, SR70415 [REF:609618455])]
Version 9.1.3.16 QL 3.03.21 4.00.23
* This version replaced the invalid release 9.1.3.15. * Firmware versions: 3.03.21 (2Gb), 4.00.23 (4Gb). * API versions: 1.28.0.35 (QLSDM), 2.0.0.14 (QL2XHAI2). * Removed I/O Reduction support for the 4G Host Bus Adapter.
Version 9.1.3.15 QL 3.03.21 4.00.23
* Firmware versions: 3.03.21 (2Gb), 4.00.23 (4Gb). * API versions: 1.28.0.35 (QLSDM), 2.0.0.14 (QL2XHAI2). * Added OEM specific support for assigning a virtual world wide port name via registry key. * Added "Get/Set" target link speed support. * Changed the INF to enable the I/O Reduction for the 4Gb Host Bus Adapter for the IA64 platform based on Microsoft document. * Added multiple BIOS support in QLSDM. * Initialized debug file name variable during DLL load (QL2XHAI2). * Closed EMC OPT23605: x64 Server hangs during boot at the “Applying Computer Settings.” * Cleaned up code related to get Object text for OS Device Name (QL2XHAI2). * Attached target devices not visible to OS after server reboot (QLE220) [ER47338 closed]. * Driver not Login into Target Devices (QLE220) [ER47436 closed] * Topology diagram is not displaying properly when LIP issued on QLA2462 port. [ER47530 closed]
Version 9.1.2.19 QL 3.03.19 4.00.23
* Firmware versions: 3.03.19 (2Gb), 4.00.23 (4Gb). * Bus Resets Under Heavy I/O Load. [ER46891 closed] * API Versions: 1.28.0.33 (QLSDM), 2.0.0.12 (QL2XHAI2). * SANsurfer FC HBA Manager B83 showed only one port see the storage even though no zoning. [ER47166 closed] * Port failure after disconnecting and reconnecting ISL port used by MirrorView. [Closed OPT 235681] * SANsurfer FC HBA Manager does not display the topology view correctly when the Host Bus Adapters are set to loop only until the system is rebooted. [ER47235 closed]
Version 9.1.2.18 QL 3.03.19 4.00.22
* Firmware versions: 3.03.19 (2Gb), 4.00.22 (4Gb). * API Versions: 1.28.0.33 (QLSDM), 2.0.0.12 (QL2XHAI2). * Added VPD update support for 2432 4-port adapter in 1.28.0.33 (QLSDM). * Recognized Asynchronous Event 8048h as a valid AEN. * Modified timer task to handle the case when Host Bus Adapter is in a stopped state. * 24xx only: Added support for NVRAM Nodename option bit. * 4Gb only: Added support for reading data from SFP (small form-factor pluggable) optical transceivers via QLSDM. * Adhered to Subsystem Id document v2.2g. * Added support for the following devices (supported platforms): - PCI\VEN_1077&DEV_2422&SUBSYS_014D1077 (x86, IA64, x64) - PCI\VEN_1077&DEV_2432&SUBSYS_014C1077 (x86, IA64, x64) - PCI\VEN_1077&DEV_2432&SUBSYS_01521077 (x86, IA64, x64) - PCI\VEN_1077&DEV_2432&SUBSYS_01531077 (x86, IA64, x64) - PCI\VEN_1077&DEV_2432&SUBSYS_01541077 (x86, IA64, x64) - PCI\VEN_1077&DEV_5432&SUBSYS_014E1077 (x86, x64) * Adhered to firmware dump procedure per 2400 Series Firmware Interface Specification version C. * RX2600 fails to boot Windows after 1MB bin flash—SANsurfer 2.0.30b84_windows. [ER47113 closed] * Fixed: LUNs disappeared after DS4000 controller firmware download in direct attach. * "Bind By Port ID + Present Targets that are persistently bound only" is not working (in loop environment). [ER46586 closed] * Changed to perform the full firmware dump when firmware error occurs. * JBOD disks turn offline in SANsurfer FC HBA Manager and SANsurfer FC HBA CLI when settings are Present targets that are persistently bound + Bind by Port ID (in loop environment). [ER46141 closed] * Corrected firmware region size in option ROM layout, that is, firmware size in flash. * GPN_ID has wrong payload length after unplug and re-plug. [ER45035 closed] * When using the QLSDM to retrieve the transceiver data, you may get SDMGT_CANNOT_GET_SFP_DATA error on system platforms using specific models of Supermicro motherboard such as P3TDL3 and P3TDLE. The QLSDM retrieves the transceiver data using underlying 4Gb firmware (v4.00.21) that reports an error code of 4005 for this operation. * Verified support for Atmel flash chip AT26DF081A.
Version 9.1.2.16 QL 3.03.19 4.00.18
* Firmware versions: 3.03.19 (2Gb), 4.00.18 (4Gb). * Adhered to Subsystem Id document v2.2f. * Added support for the following devices (supported platforms): - PCI\VEN_1077&DEV_6312&SUBSYS_01191077 (x86, x64) - PCI\VEN_1077&DEV_6312&SUBSYS_011C1077 (x86, x64) - PCI\VEN_1077&DEV_6312&SUBSYS_01301077 (x86, x64) - PCI\VEN_1077&DEV_6322&SUBSYS_012F1077 (x86, x64) * Set additional firmware option bit 10 for 2Gb Host Bus Adapter so the firmware waits for the task management command, (such as reset LUN to complete.) * Host Bus Adapter does not re-login to target after target sends LOGO to Host Bus Adapter. [Closed OPT226506] * The link status behaves differently between QLE2460 and QLA2342. [ER44735 closed] * Incorrect HwVersion string is reported by QLogic IHV HBA_GetAdapterAttributes call. [ER45228 closed] * Event Id 11 entries logged by ql2300 at offset x10 and offset x34 do not match the Error Codes published on the Web. Because the miniport uses Microsoft's Storport.sys driver to actually log the error event, and it has been confirmed that the storport.sys driver logs the event incorrectly. A bug has been reported to Microsoft on this: SRX041110604850 (STORport—Incorrect event data is logged when storport's storportlogerror). Meanwhile, the Miniport driver is changed to log the event code at offset x10 by default. This workaround is controlled by the registry parameter mevtwrkaround. For details on this parameter, please see the readme file. [ER45230 closed] * Use memory map mechanism in the driver. [ER45232 closed]
Version 9.1.2.15 QL 3.03.19 4.00.17
* Firmware versions: 3.03.19 (2Gb), 4.00.17 (4Gb). * Adhered to Subsystem Id document v2.2e. * Added support for the following devices (supported platforms): - PCI\VEN_1077&DEV_5432&SUBSYS_013E1077 (x86, IA64, x64) - PCI\VEN_1077&DEV_5432&SUBSYS_013F1077 (x86, IA64, x64) - PCI\VEN_1077&DEV_2422&SUBSYS_014A1077 (x86, IA64, x64) - PCI\VEN_1077&DEV_2432&SUBSYS_014B1077 (x86, IA64, x64) * Added support to use combined 2Gb IPX firmware. * Added support for SST25VF080B flash part. * Added option for not to dump code section for 24xx firmware dump. * Serial number for the 4Gb Host Bus Adapter should be obtained from the VPD. [ER44270 closed] * Windows 2000/qla2340/Veritas tape backup long cable pull failed error message ERROR_DEVICE_NOT_CONNECTED. [ER38787 closed]
Version 9.1.1.15 QL 3.03.18 4.00.12
* Firmware versions: 3.03.18 (2Gb), 4.00.12 (4Gb). * QLogic 4Gb driver returns selection timeout immediately after cable pull in an OEM special dual loop configuration. [Closed PTR 90818] * QLogic Driver returns a selection timeout when rebooting a controller in an OEM special dual loop configuration. [Closed PTR 90652]
Version 9.1.0.18 QL 3.03.18 4.00.12
* Firmware versions: 3.03.18 (2Gb), 4.00.12 (4Gb). * Adhered to Subsystem Id document v2.2d. * Added support for the following devices (supported platforms): - PCI\VEN_1077&DEV_2422&SUBSYS_12D7103C (x86, IA64, x64) - PCI\VEN_1077&DEV_2312&SUBSYS_01491077 (x86, IA64, x64)
* Firmware versions: 3.03.16 (2Gb), 4.00.12 (4Gb). * Embedded 24xx firmware in the driver. * New driver could not detect any device to install as BOOT from SAN on one specific setup. [ER43175 closed]
Version 9.1.0.15 QL 3.03.16
* Firmware versions: 3.03.16 (2Gb), (Use onboard 4Gb firmware.) * Added 24xx support. * Need to change bit used to control LED mode. [ER42601 closed]
Version 9.0.2.17 QL 3.03.13
* Firmware versions: 3.03.16 (2Gb).
Firmware (Embedded in Driver) -----------------------------
The following lists show the changes and bug fixes to the firmware that is embedded in the drivers for the 4Gb Fibre Channel Adapters and 8Gb Fibre Channel Adapters from versions 4.03.01 to 5.04.04.
From Version 5.04.02 to 5.04.04
Changes ------- * 2500 Series. Ensured that correct slot-specific SerDes settings will be used for specific mezzanine cards. [ER78047]
Bug Fixes --------- None
From Version 5.04.01 to 5.04.02
Changes ------- * Current Fibre Channel Adapter firmware supports maximum frame sizes of 512, 1024, and 2048 in the received PLOGI or PLOGI-ACC. This change provides an option to disable this restriction by setting bit 6 of Firmware Options 2 through Set Additional Firmware Options Mailbox Command (038h). Note that this option bypasses the Maximum Transmit Data Frame Size set by the Set Transmit Parameters Mailbox Command (019h). [ER76448]
Bug Fixes --------- None
From Version 5.04.00 to 5.04.01
Changes ------- * Option to Disable Automatic FDISC: Currently, the firmware automatically performs FDISC for enabled virtual ports (VPs) upon receiving an FLOGI_ACC. This procedure does not mesh well with driver architectures, which typically control the FDISCs. A new option bit allows you to suppress the automatic FDISCs. This option bit is available in both the ISP25xx and ISP81xx firmware. Older firmware ignores this option bit. Functional Description: The Disable FDISC bit (multi-ID initialize firmware control block, Global VP Options field, bit 4) disables the automatic FDISCs. If this bit is set, any VPs enabled in the multi-ID initialize firmware control block remain enabled, but the ID is not in the acquired state. The driver must issue a Modify Virtual Port Configuration IOCB (with the Command field set to 3h) to instruct the firmware to transmit the FDISC after it receives the report ID acquisition (RIDA) for VP0. When Disable FDISC is set, no FDISC is transmitted for the enabled VPs on any event, which results in a FLOGI. The driver should issue a Modify Virtual Port Configuration IOCB (with the Command field set to 3h) to instruct the firmware to transmit the FDISC for a specific VP (after receiving the RIDA for VP0). The driver must instruct the firmware to transmit an FDISC for all enabled VPs. After all the FDISC is completed, the firmware state changes to ready. In addition, a Port Database Changed AEN (8014h), with the Outgoing Mailbox 1 register set to FFFFh and Outgoing Mailbox 2 register set to 6h, is generated and passed to the driver. If the driver does not transmit FDISC for a VP, the driver must disable that specific VP using Virtual Port Control IOCB (Entry Type field 30h). This step is needed because a VP is enabled, but the ID is not in the acquired state, which can prevent the firmware state from changing to ready, and prevent generation of the Port Database AEN as previously stated. The option applies only to fabric point-to-point topology and NPIV environments. [ER74448]
* Added option to disable PRLI REC_SUPPORT bit. To disable the PRLI REC_SUPPORT bit (PRLI REC_SUPPORT bit is set by default), set bit 6 of the additional firmware option 3 of the Set Additional Firmware Options (38h) Mailbox Commands). This option has to be set prior to Initialize Firmware mailbox command (60h or 48h). [ER74764]
* 81xx only. Added Additional Firmware Options 2, bit 9 to improve NIC bandwidth when NIC and FCoE receive traffic are running concurrently. [ER69565]
* 81xx/25xx MIDQ only. Provide capability in Queue Management (1Fh) mailbox command to move a queue base address. For details, refer to the Multiple Queue Support for 2500 Series Fibre Channel and 8100 Series Converged Network Adapter Controllers Applications Note. [ER69606]
* 81xx only. Added support for LR Optics. Requires MPI FW 1.39.00 or later with proper configuration files. [ER69613]
* Added an Additional Firmware Options 3 bit to discard any frames received out of order unless the Initialize Firmware Options 3 bit 6 (Enable Receive Out-of-Order Data Frame Handling) is set to 1. Changes to Set Additional Firmware Options (38h) Mailbox Command interface: - Add to Additional Firmware Option 3 Bits (Incoming mailbox 3 register) Table: Bit 15 Do not transfer received out of order data for an exchange to host memory (except during SRR recovery). - Replace bit_15 description with: Bit 15 (Do not transfer received out of order data for an exchange to host memory). When this bit is set to 1, the ISP firmware does not transfer received out-of-order data for an exchange to host memory, except during SRR recovery. A frame is considered "out-of-order" after a previously-received frame is dropped due to an error, as well as a true out-of-order frame receipt condition. NOTE: Bit 15 is ignored when Initialize Firmware Options 3 bit 6 (Enable Receive Out-of-Order Data Frame Handling) is set to 1. [ER70311]
* 24xx/25xx only. Bit 8 of State Flags field in Status IOCB Type has been changed from Reserved to "Unable to send OPN to Destination." This bit is set when the destination device sends an immediate CLS in response to an OPN from ISP. [ER71132]
* Add option to disable the firmware automatic port login (PLOGI) to the simple name server (SNS). To enable this feature, set Bit 15 of Additional Firmware Options 2 when issuing Set Additional Firmware Options Mailbox Command 0038h. [ER71742]
* 24xx SB2 only. Return the lower 16-bit of current RBC count in Receive IU and Add Buffer IOCBs. This extends the scope of ER68875 to include 24xx. - For IOCB type 0x4D, the lower 16-bit of the counter is stored at offset 0x3E to 0x3F. - For IOCB type 0x4C, the lower 16-bit is stored at offset 0x1C to 0x1D. [ER71764]
* 81xx only. Added an Additional Firmware Options 2 bit to post all Asynchronous IOCBs to the ATIO queue when target mode is enabled. Changes to Set Additional Firmware Options (38h) mailbox command interface: - Add to Additional Firmware Option 2 Bits (Incoming mailbox 2 register) Table: Bit 11 Enable Asynchronous IOCBs to ATIO queue. - Replace bit 11 description with: Bit 11 (Enable Asynchronous IOCBs to ATIO queue). When this bit is set to 1, the firmware posts all asynchronous IOCBs (PUREX, ABTS Received, and Report ID Acquisition) to the ATIO queue when target operation is enabled. Additionally, if a login-affecting ELS command (PLOGI, PRLI, PRLO, LOGO, ADISC, PDISC, or TPRLO) is passed up to the target driver, an Immediate Notify IOCB with Status Code 46h, Login affecting ELS command received, shall not be sent. Initialize Firmware Options 3 bit 26 (Enable Ethernet Header Receive IOCB for ATIO Queue) is used to request the firmware place an Ethernet Header Receive IOCB on the ATIO queue before each PUREX and ABTS Received IOCB. Initialize Firmware Options 3 bit 27 (Enable Ethernet Header Receive IOCB for Response Queue) shall be ignored. When bit 11 is set to 0, the firmware posts asynchronous IOCBs to queues specified in the ISP81xx firmware specification. Changes to Initialize Firmware, Firmware Options 3: - Bit 27 (Enable Ethernet Header Receive IOCB for Response Queue). When Additional Firmware Option 2 bit 11 (Enable Asynchronous IOCBs to ATIO queue) is set, this bit is ignored. - Bit 26 (Enable Ethernet Header Receive IOCB for ATIO Queue). When Additional Firmware Option 2 bit 11 (Enable Asynchronous IOCBs to ATIO queue) is set, and this bit is set, the firmware places an Ethernet Header Receive IOCB on the ATIO queue before each PUREX and ABTS Received IOCB. [ER71877]
* MID only. Send LOGO if an FDISC is rejected with error/explanation code 03/0D. After LOGO ACC is received, the firmware retries the FDISC. [ER72460]
Bug Fixes --------- * 81xx only. Corrected FCP-2 processing during a link disruption until re-discovery is complete. [ER66337]
* MIDQ builds. Corrected N_Port-N_Port FLOGI Dest ID of 0. [ER67952]
* 2532-based adapter only. Corrected an issue that could prevent the link from being established when connected to specific targets operating at 8Gb link speed in a loop topology. [ER68658]
* Correct abort handling of a Command Type 6 IO that could result in fatal firmware error or I/O timeout. [ER68965]
* 81xx only. Set a Buffer Credit value of 16 in FLOGI/FDISC for interoperability with FCFs that incorrectly use the field. [ER69182]
* 81xx only. Correct handling of received stale frames forwarded by an FCF after the link is re-established. [ER69245]
* 81xx only. Corrected an issue that would prevent the link from being re-established after a cable pull. [ER69291]
* Correctly set the "sequence initiative transferred bit" (F_CTL) for transmission of status mode 2 FCP_RSP with confirmation request. [ER69336]
* 81xx only. Added support to detect MPI pause during heartbeat processing. [ER69457]
* Disk CRC only. Correct firmware reported residual Byte Count for CRC (T10) Read commands. [ER69564]
* 81xx only. Corrected CTIO handling when the link was down. [ER69683]
* 81xx only. Corrected an issue with Internal Loopback failure reported when link is up to an FCF. [ER69685]
* 24xx CRC only. Resolve 8002 system error during DIF error processing. [ER69922]
* 25xx and 81xx only. Corrected possible memory leak during clean-up of T10 CRC IOs with DIF Bundling enabled. [ER70018]
* 25xx and 81xx only. Corrected false DIF error reported during T10 CRC transmit IO with DIF Bundling. [ER70038]
* Corrected Target XCB count returned by Get Resource Counts Mailbox Command (0042h). [ER70149]
* Corrected a RISC pause condition after the Diagnostic Echo Test (44h) mailbox command completed with an error status. [ER70166]
* 25xx and 81xx only. Corrected a potential I/O timeout of T10 CRC IO when using DIF bundling. [ER70319]
* 25xx and 81xx only. Corrected potential invalid PCIe access when aborting a T10 CRC IO when using DIF bundling. [ER70558]
* Corrected N_Port Handle assignment by firmware when Purex is enabled. [ER70575]
* 81xx only. The firmware could incorrectly set bit 15 of N_Port Handle returned by the Get Port/Node Name List (75h) mailbox command. [ER70664]
* 81xx only. Corrected a problem where the values supplied in the Set Timeout Parameters (32h) mailbox command were not honored. [ER70708]
* Prevent 8002 system error when processing a severely malformed IOCB. [ER70910]
* Corrected a RISC pause condition during SRR received notify ACK processing. [ER70921]
* 24xx only: Improved link init time when using Inline Credit Extender in an N_Port-N_Port configuration. [ER71034]
* Fixed Loopback mailbox command timeout when Fibre Channel event (FCE) extended tracing is enabled. [ER71269]
* Corrected a problem that could result in a Task Management command timeout. [ER72100]
* Corrected handling of frames received with DID = 000000h. [ER72270]
* 81xx only. Fix CTIO with terminate exchange bit returned good completion status (01h) after the CTIO is already completed. The CTIO with a terminate exchange bit should instead return an invalid receive exchange address (08h) completion status. [ER72491]
* MID builds only. Fix CTIO with terminate exchange bit failed with completion status=29h. [ER72571]
* 81xx only. Corrected a problem that caused FDISC not generated by the firmware when Virtual Port Control IOCB (type=30h) is used to enable VP. [ER72921]
* MID only. Correct the cleanup of outstanding ELS requests during VP disable. [ER73048]
* MID only. Resolve RISC pause after cable pull or link event. [ER73090]
From Version 5.02.00 to 5.03.00
Changes ------- * Added feature to return LOGIN info via Mailbox Command 5Ah - Get Params, using type 7. Entry Parameters (Incoming): Mailbox0 - opcode (005ah) Mailbox1 - Type<15:8>, reserved<7:0> Mailbox2 - buffer addr<31:16> Mailbox3 - buffer addr<15:0> Mailbox6 - buffer addr<63:48> Mailbox7 - buffer addr<47:32> Mailbox8 - DW Xfr Length (only valid for Type 7) Exit Parameters (Outgoing): Mailbox0 - (4000h) normal completion - (4002h) DMA error - (4005h) error during command execution - (4006h) if unsupported type or DW Xfr requested is too small ( 28DW <.. ) data got truncated Mailbox1 - If Outgoing Mailbox 0 is 4005h, then Outgoing Mailbox 1 will have the following sub-code: 0002h - No IOCB Mailbox8 - DW Xfr Data (should be 28 for typical PLOGI, only valid for Type 7)
The format returned is according to FC-LS-2 (Fibre Channel - Link Services - 2), rev.2.00 - FLOGI/PLOGI payload minus ELS command code. [ER66811]
* 8100 only. Added support for MPI heartbeat feature. Contact QLogic for details. [ER67656]
* 8100 only. Changes to support additional MPI mailbox commands in the range 136h to 150h. [ER67716]
* 2532/8100 CRC. Implemented CTIO Type CRC_2 and support DIF bundling on CTIO CRC IOCBs.[ER67727]
* 2532 SB2 only. Return the lower 16-bit of current RBC count in Receive IU and Add Buffer IOCBs. - For IOCB type 0x4D, the lower 16-bit of the counter is stored at offset 0x3E to 0x3F. - For IOCB type 0x4C, the lower 16-bit is stored at offset 0x1C to 0x1D. NOTE: This feature will be documented in the upcoming SB-4 spec. [ER68875]
Bug Fixes --------- * MID only. Do not report RSCN for VPs that have been disabled. Previously, the firmware generated 8015 AEN with an invalid vpindex. [ER62856]
* 2532/8100 CRC T10. Resolved bad FCP_DATA sequence due to incorrect re-initialization of the CRC context for Header mode. This affected T10 write commands with multiple FCP_XFER_RDYs. [ER67220]
* Removed zeroed optional association headers from RRQ payload. [ER67210]
* 24xx/2532 only. Corrected a problem in PLOGI ACC processing that could result in failure to respond to a subsequent PRLI. [ER67333]
* 8100 only. Changed login/logout IOCB processing to correctly format a LOGO sent to the fabric controller. [ER69214]
* Resolved a Fatal Firmware Error (8002h) due to a race condition when the exchange timer expires at the same time FCP_XFER_RDY is received. [ER69068]
* 8100 only. Changed link down handling to prevent a stall in received FCoE frame processing. [ER68914]
* Prevented Fatal Firmware Error (8002h) when processing an invalid ELS response type. This problem could occur if a switch sent an erroneous ELS response to a CT pass-thru request. [ER68488]
* 8100 only. Fixed a hang that could occur when processing a new request for an FCP2 capable device after a link status change. [ER68339]
* 2532/8100 only. Resolved intermittent Fatal Firmware Error (8002h) and FKA failures during Flash access. [ER68257]
* Modified RNID processing to correctly handle a response with a "Specific Identification Data Length" field of zero. [ER67623]
* 2532/8100 CRC T10. Changed buffer pointer validation to prevent erroneous 8002 system errors. [ER67568]
* 24xx/2532 only. Fixed a Fatal Firmware Error (8002h) caused by a problem cleaning up after a port discovery timeout. [ER67345]
* 8100 only. Prevented accidental overwrite of the PortName within a port descriptor data structure. Problem could occur during the execution of mailbox commands 30h, 31h, 51h, or 60h. [ER67181]
* Fixed 200ms delay for Command Type 6 IOCBs when residual underrun occurs, and FCP_RSP DMA is enabled. [ER67159]
* 8100 only. Resolved intermittent resource allocation failures in Read SFP mailbox command (31h). Improved failure sub code reporting in mailbox 1 for both Read(31h) and Write(30h) mailbox commands. [ER69146]
* Corrected ADISC ELS check for ELS passup and fixed potential bad pointer when building ELS frame. [ER67799]
* Fixed mailbox command Diagnostic Loop Back failures with Status 4005 and Subcode 0x17. [ER67690 8100 only]
* Fix erroneous discard of a valid FIP VLAN Notification frame when the frame does not have an IEEE 801.2 Q Tag and the FIP descriptor list length is greater than 8 (words). [ER67062 8100 only]
* 24xx/2532 only. Improved firmware handling of I2C SFP interface to better support some SFPs. [ER67479]
* 8100 only. Report NIC driver attached status in Get firmware State Mailbox Command (069h). For details, see the 8100 Series Fibre Channel Firmware Interface Specification, rev C. [ER65051]
* 8100 only. Firmware will return parameter error (4006h) if the Initialize Firmware Mailbox Command (060h) or Initialize Firmware Multiple IDs Mailbox Command (048h) specifies use of the optional Extended Initialization Control Block and provide a length of zero bytes. [ER65184]
* 8100 only. Changed the Perform Implicit LOGO Mailbox Command (072h) to allow the Perform Logout for All Ports across All FCFs option (bit 1 of incoming Mailbox Register 1) in single FCF configurations. [ER65220]
* Added a feature to reject PRLI from Fabric Control (FFFCxx) when target mode is disabled. To enable this option, set bit 7 of Mailbox 2 in the Set Additional Firmware Options Mailbox Command (0038h). [ER66061]
* 8100 only. Changed Get Link Status/Read Port Status Mailbox Command (06Bh) and Get Link Statistics and Private Data Mailbox Command (06Dh) to return counters consistent with the FCoE Link Error Status Block (LESB). Figure 3-4 of the 8100 Series Fibre Channel Firmware Interface Specification now contains the following counters: Link Failure Count, Virtual Link Failure Count, Missing FIP Keep Alive or Discovery Advertisement Count, Symbol Error During Carrier Count, Error'd Block Count, and Frame Check Sequence Error Count. [ER66125]
Bug Fixes --------- * MID only. Fixed an issue to prevent invalid VP misconfigurations that can result in a Fatal Firmware Error 8002h. [ER63706]
* 8100 only. Corrected ELS and BLS timeout units conversion when processing Mailbox Commands Get Timeout Parameters (0022h) and Set Timeout Parameters (0032h). [ER65155]
* 8100 only. Corrected an issue after External Loopback was performed that prevented FIP from completing. [ER65165]
* SB2 only. Corrected a condition in RRQ processing that could result in a Fatal Firmware Error (8002h) after receipt of an ABTS. [ER65232]
* Fixed an issue that caused a rediscovery of FCFs if FIP FLOGI timed out. [ER65757]
* Corrected a problem in target mode operation during confirmation processing that could result in a Fatal Firmware Error 8002h. [ER65415 8100 only]
* 24xx and 2532 only. Corrected auto-negotiation algorithm that would restore the link at the previous speed instead of the maximum speed. [ER65802]
* 8100 non MID only. Corrected a condition where the firmware state would not become Ready after switch port was shutdown and restored. [ER65980]
* Retry FIP VLAN Discovery after FIP Discovery Solicitation fails to find a primary FCF. [ER65980]
* Return LoginState in IO Parameter 1 when Login/Logout IOCB PRLI IO Parameter 0 = 4 (Command Failed). [ER66045]
* 2532 and 8100 MID only. Corrected VP Index range checking during processing of VP Control IOCB. [ER66068]
* Corrected Firmware Error reported during Purex IOCB processing if the Port Database for the target device had been initialized. [ER66082]
* 8100 only. Fixed detection of EP device. [ER66204]
* MID only. Corrected a problem which could result in LOGO sent with SID=000000 or incorrect reporting of VP status after a VP is disabled. [ER66553]
* 2532/8100 MIDQ only. Resolved a long delay in processing Command Type 6 IOCBs after link level fault injection. [ER66755]
* Fixed RISC pause condition when using T10 CRC with DIF_bundling. [ER66803]
* 2532 and 8100 only. Add DSD chaining support to command CRC type IOCBs. [ER62531]
* 8100 only. Updated Immediate Notify to reflect format described in the firmware spec. For details, see the 8100 Series Fibre Channel Firmware Interface Specification Rev C. [ER63872, ER64455]
* 8100 only. When detected, report MPI configuration error and reason code in Get Firmware State Mailbox Command (069h). For details, see the 8100 Series Fibre Channel Firmware Interface Specification Rev C. [ER64342]
* 8100 only. Added new error completion codes to Restart MPI Mailbox Command (03Dh). For details, see the 8100 Series Fibre Channel Firmware Interface Specification Rev C. [ER64391]
* 8100 only. Implement SFP Read/Write Mailbox Commands (031h and 030h). For details, see the 8100 Series Fibre Channel Firmware Interface Specification Rev C. [ER64406]
* 8100 only. Modify Report ID Acquisition IOCB to return subcode for status 0x01, Unable to acquire ID, for FDISC. The subcode is returned at offset 10h. [ER64497]
* 8100 only. Update Get DCBX Parameters Mailbox Command (051h) to match latest spec. For details, see the 8100 Series Fibre Channel Firmware Interface Specification Rev C. [ER64672]
* 8100 only. Implement Get XGMAC Statistics Mailbox Command (07Ah). For details, see the 8100 Series Fibre Channel Firmware Interface Specification Rev C. [ER64673]
* 8100 only. Return SFP status information. See Asynchronous Event 8012h and Mailbox Commands 05Dh, 048h, 060h, and 069h. For details, see the 8100 Series Fibre Channel Firmware Interface Specification Rev C. [ER64732]
Bug Fixes --------- * 8100 only. Corrected handling of FLOGI Timeout and Reject processing by using Retry Count and Timeout as specified in Initialize Firmware Control Block. [ER63856]
* Corrected a condition where an Immediate Notify could be sent to the system driver when a VP is not configured for Target Mode operation. [ER64176]
* Corrected cleanup of pending Transmit Data requests across fault injection. [ER64251]
* 8100 only. Corrected a potential firmware hang condition after processing an ELS timeout or recovering from an internal error. [ER64265]
* SB2 only. Corrected a condition where the System Defined 1 field of an IOCB being returned to the system driver may be overwritten. [ER64315]
* 8100 only. Correct an issue with Flash Access Mailbox Command (003Eh). [ER64377]
* 8100 only. Corrected potential FCoE traffic stall after Port Reset completes. [ER64435]
* 8100 only. Fix FIP Failure to find FCF after a number of cable pulls/insertions into different FCF ports. [ER64454]
* 8100 only. Get FCF List Mailbox Command (050h) may fail to find valid FCF Index. [ER64507]
* 8100 only. Corrected false Loopback errors reported during internal and external tests. [ER64520]
* Corrected potential RISC pause when PUREX is enabled. [ER64520]
* 8100 only. Fixed incorrect Destination MAC address in FIP FLOGI request after an FCF Fabric name change. [ER64585]
* 8100 only. Allow primary port to be discovered via FDISC after it received a Clear Virtual Link or Fabric LOGO and another Virtual Port becomes the master. [ER64590]
* 2400/2532 only. Corrected a condition where a FCP RSP could be transmitted across a LIP and before the discovery timer expires in private loop topology. [ER64705]
* 8100 only. Corrected Firmware hang condition after receiving an ABTS. [ER64902]
From Version 4.06.00 to 5.00.00
Changes ------- * MID only. The firmware now returns a Port Unavailable (28h) Completion Status if the VP Index fails validation during a SCSI Task Management request. [ER62857]
* MID only. The firmware now allows Get VP Database Mailbox Command (49h) regardless of VP Status. [ER62858]
* SB2 only. The firmware now allows Get Link Status Mailbox Command (06bh) to complete even though Process Login has not completed. [ER63650]
Bug Fixes --------- * Fixed an issue to prevent initiating an LR upon a Transmit timeout while the firmware services other requests. [ER61666]
* Fixed an issue that delayed starting new Transmit requests if a Transmit error is pending. [ER61682]
* Corrected the auto-negotiation algorithm that could prevent the link from establishing when connected to a switch. [ER62000, ER62172]
* Corrected CTIO handling when link is down and CTIO received with an Exchange Address of FFFFFFFFh. [ER62715]
* MID only. Corrected various issues relating to VF support. [ER62798]
* Corrected issues found with VF and Fabric LOGO processing. Also corrected the valid VP Count range to 1-254. [ER62826]
* MID only. Corrected Source and Destination Address on an SCR retry. [ER62886]
* Completed various FCE tracing fixes. [ER62952, ER63400]
* Customer-specific fix. [ER63104]
* Corrected IDLE character to ArbFF when connected at 8Gb rate. [ER63488]
* Corrected an issue where a Mailbox or Mailbox IOCB request would not complete if a Fabric LOGO occurred. [ER63563]
* DISK CRC only. Corrected a RISC Pause condition reported during cable pull testing.[ER63564]
* Fixed an incorrect Response In Pointer reported in the RISC to Host Status Register when running in ZIO mode. [ER63790]
* 2532 MID multiple queue only. Corrected a link credit issue that resulted in decreased performance. [ER63913]
From Version 4.05.00 to 4.06.00
Changes ------- * 2532 only. Added support for Virtual Fabrics. Contact QLogic for API changes to support VFs. [ER60036, ER60250]
* MID only. Added feature to allow the firmware to perform SNS Login and SCR registration for virtual ports. This feature is enabled by setting bit 6 of the Options field during Initialize Firmware - Multiple IDs Mailbox Command (048h). Added the following status/ substatus values to Report ID Acquired IOCB:
VP Status Description --------- ----------- 3 SNS Timeout. See byte offsets 23:20 for Sub-status. 4 SNS LS_RJT (Reason/Explanation/Vendor Unique in IOCB byte offsets 23:20). 5 SCR Timeout. See byte offset 23:20 for Sub-status. 6 SCR LS_RJT (Reason/Explanation/Vendor Unique in IOCB byte offsets 23:20).[ER61930]
Bug Fixes --------- * MID only. Process received LOGO from F_Port. The associated Port Database Changed (8014h) asynchronous event will differ from a LOGO received from an N_Port as follows:
* 2532 only. Added the ability to specify a maximum Fibre Channel Link speed to be negotiated. This feature is enabled by bit 12 of the Initialize Firmware Control Block’s Firmware Option 3. When this bit is set to 1, firmware will negotiate the FC data rate from the given max speed specified by driver in bits 15-13 of the Initialize Firmware Control Block’s Firmware Options 3 field to two speeds down to what the SFP allowed. This option can also be set in the Data Rate Mailbox Command (05dh). When incoming mailbox 1 is set to 03h, firmware forces a loss of sync, restarts link initialization, and negotiates the FC data rate from the given max speed specified by driver in incoming mailbox 2 to two speeds down to what the SFP allowed. [ER60321]
* Allow system driver to send LOGO ELS via ELS Pass-Through IOCB using an N_Port Handle of FFFFh. [ER60400]
* MID only. Return Command Parameter Error (4006h) if the number of requested VP’s exceeds the limit when issuing Initialize Firmware – Multiple-IDs Mailbox Command (048h).
* 2432 only. Disable PCIe Completion Timeout. PCIe Completion Timeout has been disabled to address issues seen on certain systems where PCIe completion could take over 50ms. [ER61461]
Bug Fixes --------- * MID only. Corrected problems related to fault injection. [ER59740, ER60612]
* MID only. Various fixes to improve detection and processing of topology changes. [ER60127, ER60326, ER60610, ER60633]
* Corrected timing of when RRQ is issued after ABTS ACC is received. [ER60248]
* Corrected Out-of-Order handling of overlapped or repeating received XFR_RDY frames. [ER60318]
* 2532 Disk CRC only. Various Disk CRC changes to remove Fatal Firmware errors (8002h) being reported. [ER60336, ER61222, ER61223]
* MID only. Corrected a problem where a LOGO ELS could be sent to a port that was not logged in. [ER60565]
* MID only. Corrected a problem where a FLOGI/FDISC would not be sent to the switch after a previous FLOGI/FDISC completed with an ELS RJT. [ER60566]
* MID only. A PLOGI could be rejected for a Virtual Port even though the Primary Port parameters have not changed. [ER60568]
* MID only. Corrected a problem where an incorrect number of acquired AL_PAs could be reported during the LIRP and LILP phases of loop initialization. [ER60569]
* MID only. Login/Logout IOCB completion status was incorrect when no FLOGI ACC ELS had been received. [ER60571]
* Corrected a Fatal Firmware Error (8002h) when a CTIO IOCB with Confirmation Satisfied is received when the FCP_RSP has not been sent yet. The CTIO will be returned with a completion status of Invalid Request (06h). [ER60572]
* MID only. Corrected handling of incoming frames while waiting for LOGI ACC ELS for the primary port. [ER60576]
* Corrected a problem when attempting to abort an ELS Pass-Through operation when a PLOGI is received that changes the ports D_ID. [ER60637]
* Class 2 only. Corrected Fatal Firmware Error (8002h) reported when attempting to abort transmit operations. [ER60698]
* Corrected timer values used with link recovery. [ER60877]
* Improved handling of receiving an ABTS for a CT Pass-Through exchange. [ER61007]
* Corrected a problem with aborting IOs for the incorrect LUN when a LUN Level Task Management Function is active. [ER61155]
* MID only. Ensure that Buffer-Buffer Credits are not modified when processing FDISC or FDISC-ACC. [ER61187, ER61216]
* MID only. Correctly track changing of local 24-bit ID during cleanup. [ER61200]
* Handle race condition of completing transmit of PLOGI-ACC and receiving PRLI. [ER61465]
* Corrected failures reported by Loopback Mailbox Command (45h) and Echo Mailbox Command (44h). Added sub-error codes to Outgoing Mailbox 3 of the Echo Command to match the Loopback Command. [ER59630, ER59631]
Bug Fixes --------- * Disk CRC only. Corrected handling of frames received out of order in target mode. [ER58251, ER58760, ER58761]
* 2532 Disk CRC only. Corrected condition where the firmware would overwrite the IOCB Entry Status field when the device receives invalid IOCBs. [ER58532]
* 2532 Disk CRC only. Fixed a condition that could result in false DIF Errors when receiving runt block data. [ER58533]
* Disk CRC and 2532 only. Corrected an issue with multi-sequence DIF Remove functionality. [ER58752]
* 2532 only. Corrected a RISC pause that occurred when optional external memory is present. [ER58783]
* Corrected endianess of RNID user defined data field. [ER58802]
* Set correct Immediate Notify IOCB sub-code during fault injection. [ER58897]
* Improved handling of switches that transition link from F-FL-F when the device is directly connected to a switch port. [ER59100]
* Corrected issues relating to the EHAFT feature. [ER59279, ER59431]
* Corrected a target mode issue where an ATIO Queue update interrupt would not be posted when using a small ATIO queue. [ER59385]
* Corrected a Fatal Firmware Error (8002h) reported when ISP receives FCP_RSP with a payload length greater than 240 bytes. This problem would occur when Command Type 6 IOCB is used with the option to DMA the FCP_RSP payload to the host is enabled. [ER59503]
* 2532 only. Corrected the ending memory address reported by About Firmware Mailbox Command (08h) when external memory is present. [ER59793]
* 2532 only. Corrected the Report Link Statistics information for Loop topology. [ER59989]
* SB2 only. Corrected a problem where the firmware was clearing the N_Port Handle field of a SB2 Send IU IOCB. [ER60017]
* Disk CRC and 2532 only. Corrected false DMA error reported when using CTIO Type 7 IOCB (Target Mode) with Status Mode 2 selected. [ER60067]
* Fixed a problem that may cause a target level Task Management Function to fail while a LUN Level Task Management Function is outstanding. [ER60126]
* MID and 2532 only. Corrected a Fatal Firmware Error (8002h) reported after adapter side fault injection. [ER60256]
* Corrected a condition where the local loop would not recover after a transmit error was encountered.[ER60301]
From Version 4.03.01 to 4.03.02
Changes ------- * 2532 only. Enabled additional FC SerDes parameters to be set via Flash for mezzanine cards. [ER59755]
Bug Fixes --------- None
From Version 4.03.00 to 4.03.01
Changes ------- * 2532 only. Added support for mezzanine adapters Added a new bit, Embedded HBA, in Flash to indicate an adapter that does not use SFP+, and transmits FC signal over a copper backplane or mid-plane. [ER58635]
Bug Fixes --------- * 2432 only. Corrected a condition where Mailbox Commands issued to Function 1 would not complete after issuing Stop Firmware Mailbox Command (14h) to Function 0. [ER58617]
2. Known Issues
None.
3. Notices
Information furnished in this document is believed to be accurate and reliable. However, QLogic Corporation assumes no responsibility for its use, nor for any infringements of patents or other rights of third parties which may result from its use. QLogic Corporation reserves the right to change product specifications at any time without notice. Applications described in this document for any of these products are only for illustrative purposes. QLogic Corporation makes no representation nor warranty that such applications are suitable for the specified use without further testing or modification. QLogic Corporation assumes no responsibility for any errors that may appear in this document.
4. Contacting Support
For further assisance, contact QLogic Technical Support at: http://support.qlogic.com
Trademarks
Accelera, Accelerating Cluster Performance, InfiniCon Systems, InfiniNIC, InfiniPath, InfiniView, QLA, QLogic, the QLogic logo, ReadyPath, SANdoctor, SANsurfer, and SilverStorm are registered trademarks of QLogic Corporation. All other brand and product names are trademarks or registered trademarks of their respective owners.
(c) Copyright 2013. All rights reserved worldwide. QLogic and the QLogic logo are registered trademarks of QLogic Corporation. All other brand and product names are trademarks or registered trademarks of their respective owners. qlgc_dd_fc_9_1_11_26_b_windows_32-64_chg
=============================================================================== Emulex NIC Device Driver for Windows ===============================================================================
Device Driver Version: 10.2.261.11 Supported On: IBM System x, BladeCenter, and Flex
Problems Fixed: - When using VMQ with Windows Server 2012 or 2012R2, the user may experience VM connectivity loss, packet drops, system hangs, inability to shutdown VMs and possible system crashes on shutdown. The complete solution also requires 10.2 based firmware.
Incremental Interoperability: - ARI (PCI-SIG specification) - RDMA over CEE (RoCE)
Known Issues: - RoCE support is limited to named applications. Contact your local IBM Sales Support for more information.
- If the installed Emulex NIC Teaming kit is the same version as the NIC Teaming kit included in this package (2.8.5-4), the installation will fail with an error code of 15611. Although this failure will be reported, the NIC driver and OCM application software were still installed successfully.
If the NIC Teaming kit doesn't need to be reinstalled, this error can safely be ignored. If the NIC Teaming kit does need to be reinstalled, the file elxdrvr-nic-teaming-2.8.5-4.exe should be extracted from the package and run manually. Instructions for extracting the contents of the package can be found in the readme. This will be resolved in a future release.
Device Driver Version: 4.6.203.1 Supported On: IBM System x, BladeCenter, and Flex
Problems Fixed: - HyperV VM's tagged VLAN's did not get an IP Address - NIC Function Properties > Status for PCI Express Link Speed says Invalid Link Speed detected. - iSCSI ports of CNA card in ESXi 4.1u3 OCM GUI (vCenter plug in) is not getting displayed
Device Driver Version: 4.4.176.2 Supported On: IBM Flex
Device Driver Version: 4.2.390.6 Supported On: IBM System x and BladeCenter
Problems Fixed: - Cleaned up inf file - TOE Offload Fails when the same neighbor is added more than once
Incremental Interoperability: - Windows 2012
Known Issues: - Switch Independent Mode not supported in a HyperV environment - Default values vary between inbox and out of box drivers. For example: Configuration section - Wake On LAN - Inbox default is ‘Disabled’ – out-of-box default is ‘Enabled’ Performance section - Maximum Number of RSS Queues - Inbox default is ‘4’ - out-of-box default is ‘8’ - Virtual Machine Queues - Inbox default is ‘Disabled’ – out-of-box default is ‘Enabled’ You may select the button to reset all driver settings to defaults after any installation.
Device Driver Version: 4.1.370.0 Supported On: IBM System x, BladeCenter, and Flex
Incremental Interoperability: - Initial release for IBM Flex
Problems Fixed: - HS23 - Sleep Stress test failed with bugcheck C4 during driver verify - TOE with RSC enabled causes Network to go down - Augmented WOL Capabilities - Driver parameter query failures - BSOD 9F during shutdown if device stops responding - RSS queues may change port during MPRestart and MPPause
Device Driver Version: 4.1.334.25 Supported On: IBM System x and BladeCenter
Problems Fixed: - Emulex: Large Receive Offload support needed in Windows NIC driver - Win WoL: Need to report WoL based on ARM reported capabilities - update the driver strings for existing cards - RSS code may access PerMessage beyond end of array - 16 queue RSS for BE3 on Windows 2008 R2 - fix unverified mode of occfg to allow customer workarounds - 16 queue RSS lockup if SRIOV enabled in registry - Standard Property Page for 1 port Windows Azure card - Timeout RQ flush to avoid bugcheck during shutdown - Change receive buffer alignment to use Windows 2008 R2 TCPIP fast path - Emulex: BSOD - Replacing BE3 card with a BE2 card after installation of new kit driver and firmware on Win 2k3 SP2 R2x64 (vNIC enabled IBM machine) - be2nd62.inf failed ChkInf - VMQ registry keys displayed in Windows 2008 - be_function_prepare_nonembedded_ioctl overwrites version field - autoi reboot fails on Win2K8 x64 when verifier enabled - PCI ID: Updates for Endeavor 2 for IBM- PCI IDs Rev 1.22 - Emulex NIC driver causes BSoD during shutdown whenever disable firewall or join domain network - Sleep Stress test failed with bugcheck C4 during driver verify - Disable TOE by default for IBM - BSOD 9F during shutdown if device stops responding
=============================================================================== Emulex iSCSI Device Driver for Windows ===============================================================================
Device Driver Version: 10.2.254.0 Supported On: IBM System x, BladeCenter, and Flex
Problems Fixed: - Cleaned up PCI ids in the inf file - Mis-match in product version between NIC and iSCSI drivers - iSCSI driver's usage of EQ/CQ/RQ doorbell formats for XE-102 and XE-104 - Lun Reset/Target Reset can hang under some errors in Windows iSCSI driver
Device Driver Version: 4.1.334.0 Supported On: IBM System x, BladeCenter, and Flex
Incremental Interoperability: - Initial release for IBM Flex
Problems Fixed: - Logout handling on boot session when multiple sessions exists to the same boot target - Add PCI ID as per Rev 1.18 for Windows iSCSI Driver - Multipath IO fails producing a BSOD - Persistent login to Max targets + 1 from WMI causes a hang - Update Nomad Device Name as per 1.19 for Windows iSCSI drivers - PCI ID: Updates for Endeavor 2 - PCI IDs Rev 1.22
Adapters Systems: IBM Flex System x880 X6 Compute Node
NOTE TO SERVICE - Reference RETAIN #N/A
Version 24.2.1-0039 ------------------- Fixes: - Fixed an issue where system would not boot into legacy mode. - Fixed an issue where default setting of VD doesn't reflect MFC's setting. - Fixed an issue where ‘Cli show all’ command fails with error msg ‘Error! App exec failed in FW PCLIError! Internal.’ - Fixed an where ServeRAID M5210 showed the wrong slot numbers in both MR and iMR, in both HII and MSM.
Version 24.2.1-0027 ------------------- -Initial release ibm_fw_sraidmr_1210e-24_2_1-0039_windows_32-64_chg
Change History - 3.11af
Package Version 3.11af.c Only change with ".c" is with file naming. 05/13/2014
Change History - 3.11af.a 03/30/2014 ======================================================================================== Firmware
* Added check invalid LUN when configuring boot target (ER105664). * Added support for P3P * Removed Execution Throttle Option From CTRL_Q
Previous versions:
Version : 7.01.00 Date : 01/09/2014
Version : 7.00.00 Date : 07/22/2013
Version : 5.10.00 Date : 06/12/2013 * Added check invalid LUN when configuring boot target * Added support for P3P * Removed Execution Throttle Option From CTRL_Q * Fix issue that cause external looback to fail (er71773). Added support for Hilda
Version : 5.07.02 Date : 10/23/2012 * Fix Task Management IOCB type 14h completion with dma error status. * Fix an issue where HBA would constantly ISP reset if switch in transparent mode but no upstream switch
Version : 5.07.00 Date : 5/17/2012 * New Feature: Firmware sends Report ID Acquired IOCB Type 32h upon Flogi failure. * New Feature: Firmware reports NportHandle, WWPN and WWNN of remote port in Report ID Acquired IOCB Type 32h for N2N connection topology. * Fix firmware not logging out SNS port (PID=0xFFFFFc) while processing implicit Logout all request * EP Device Only: Fix advertised BB_Credits not increasing with additional receive credits option set. * Target Mode: Fix firmware fatal error seen while aborting an exchange. * Fix firmware failing request to disable multiple virtual ports using iocb type 30h. * Fix Loop port bypass (40h) and loop port enable (41h) mailbox commands not working correctly. * FC-VI only: Fix disconnect request iocb timeout. * T10 Only: Fix firmware reporting DIF errors while running multi sequence, T10-DIF enabled IO’s with protection mode set to DIF- Remove. * Fix firmware fatal error while running T10-DIF enabled IO’s with protection mode set to DIF-Remove. * Fix incorrect reporting of discarded and dropped frame count in mailbox command 6Dh. * Add support for programming page and payload length fields in TPRLO response frame.
Version : 5.06.05 Date : 4/26/2012 --Initial Release
Version : 2.14 Date : 05/07/2013 *Added check invalid LUN when configuring boot target (ER105664). *Added support for P3P *Fix issue that cause external looback to fail (er71773)
Version : 2.13 Date : 7/31/2012 *Added code code to check for non-EDC HBA's in CTRL-Q utility. *Fixed bug in EDC update code
Version : 3.20 Date : 05/07/2013 * ER102374: Display Product-ID string in VPD area during initial selection. * ER94574: Fix stack bug in .adatper-info, found in blade hotplug test by efcode
Version : 3.19 Date : 7/31/2012 * Add check for MBAR2 (Multi-Queue memory) and create reg property if it exists. * Make validate-nvram silent. Print messages only if debug-flash?(x1000) is set. * Added dynamic sizing of the BAR registers at probe-time for reg property. * Remove dis-exp (disable-exprom) in ena-reg-acc. Causes problem on SPARC T3-1B. * Fix bugs in search of Card-Type-Lists for SSDID when flashing w/find-ilt-code. * Change oem property to QLogic or Oracle, and manufacture property to QLogic. * Added SSDID check to set-edc-config. * Stripped ISP2400 & SUN specific code using #ifdefs to reduce IBM size by .5KB * Added EDC FW area in Flash and added write-sfp and read-sfp mailbox commands * Added set-edc-config, edc-update?, and edc-present! words to nvram.fth.(IBM) * Added check-edc-load and load-edc-fw words to init-isp.fth..
=============================================================================== Emulex FCoE Device Driver for Windows ===============================================================================
Device Driver Version: 10.2.261.4 Supported On: IBM System x, BladeCenter, and Flex
Problems Fixed: - ABTS sent by tape device ignored by HBA - BSOD issue during CPU Hot replace test
Device Driver Version: 2.74.016.001 Supported On: IBM System x, BladeCenter, and Flex
Problems Fixed: - Cleaned up a variety of corner case issues to avoid BSODs - Fixed IOCTL encounter bug-chk on BE3 during NPIV tests - Fixed reporting incorrect error code when creating vport on a switch without NPIV supported - Updated the default model name for 10Gb CNAs and 16G HBAs
Device Driver Version: 2.74.009.001 Supported On: IBM System x, BladeCenter, and Flex
Problems Fixed: - Clean up non-critical legacy FC and uCNA incompatibilities - LUN masking doesn't persist after reboot - Emulex GUI crashes when closing a running Upgrade utility and re-open it in 802.3 configuration with IP or MAC - Several bugs related to creation or deletion of vports
Incremental Interoperability: - Windows 2012 - Improved debug capabilities - Tighter integration with MS Windows Design Kit
Device Driver Version: 2.70.018 Supported On: IBM System x, BladeCenter, and Flex
Problems Fixed: - Fixed sending ABTS continuously to Flogi issue - Fixed reset and link bounce issues for SLI4 FC public loop - Added support for ABTS timeout handling for both SLI3 and SLI4 - Fixed loopback test incorrectly dropping the loopback frame in SLI4 - Fixed exchange management issue in SLI4 CT receive path causing array overflow - Fixed firmware version update issue following download in link up state - Fixed problem in init_vfi completion handler that stalled fabric login on BE FCoE FW - Fixed problem that caused OCM ELS Echo Test to fail - Fixed the loopback test issue in recovering from error condition - Add persistent link down support for SLI4 FC - Fix DFC_IssueDumpMBox to manage request type 3 which includes link select option - Modified reinitialization sequence following reset to fix discovery failure post type 1 or type 3 reset - Fixed new firmware dump issue
=============================================== Change History =============================================== Document Name: Change History Document Status: IBM General Avalibility Document Version: 3.2.3 Latest Driver Version: 3.2.3.0 Document Updated: 05/30/2013 Product: Brocade FC/FCoE/NIC
=================================================
Please visit www.brocade.com/hba for full documentation, additional drivers, and management applications.
Version 3.2.3.0 New Features and Feature Enhancement: - Support for patching firmware via boot image - Support for UEFI Health Check Protocol
Fixes and Enhancements: - Enhanced Windows Network performance
Version 3.2.1.0 New Features and Feature Enhancement: - Support for "IBM Flex System FC5024D 4-port 16Gb FC Adapter" PART_NUMBER "95Y2381"
Fixes and Enhancements: - Enhanced windows Hyper-V driver to avoid a infrequent crash seen during shutdown with VMQ enabled - Included windows 2012 Driver Update Disk (based on windows 2008 R2 DUD)
Version 3.2.0.0 New Features and Feature Enhancement: - BB Credit Recovery - Forward Error Correction (FEC) - Diagnostic Port (D-Port) - LUN Level IO Latency/Stats - UCM (Unified Configuration Manager) - Windows Network Performance and scalability enhancement
Fixes or enhancements: - For Brocade 8x5/4x5, I/O interrupt latency values are dynamically updated. Changes are effective for all the ports on the given adapter. - Addressed infrequent PSOD in VMware ESX with iSCSI or VLAN configured traffic in DCB enabled network. - Enhanced BR1860 Tape IO performance. - Improved BR1860 initialization on Xen kernel.
Version 3.1.0.1 Key features and enhancements supported for the software release 3.1.0.1 include the following:
New Features: - FC-AL support for direct attach to storage - Ability to assign minimum bandwidth to each vNIC - Queue Depth adjustment - FC QoS set by percentage
New Operating System Support: - Windows Server 2012
Fixes or Enhancements: - On Windows 2008 or 2008 R2, resolved an issue where HCM may freeze intermittently when Team(s) or VLAN(s) are configured. - Improved stability with Hyper-V and Teaming configurations. - On Windows 2008 or 2008 R2, resolved an issue that could stall network traffic during heavy traffic load.
Version 3.0.3.0 Initial release for 16G FC Adapter Features - Initial support for Brocade 1867 IBM 16G Mezzanine Fabric Adapter.
HBA - Fibre Channel 16G Support
Fixes and Enhancements: N/A brcd_dd_cna_3_2_4_0_windows_32-64_chg
IBM Online SAS/SATA Hard Disk Drive Update Program Version 1.14.04 Build ID: SAS11404
NOTE TO SERVICE - Reference RETAIN TIP H29851 ******************************************* IBM SAS Hard Disk Drive Update Program Version 1.14.04 Critical Release Reference RETAIN TIP H29851
This package includes updates to drives with product ID strings of:
SAS 2.5"
Initial code Release - SMK3001GRDB, SMK6001GRDB, SMK9001GRDB to SB35
SATA 2.5"
- SSDSC2BA200G3I, SSDSC2BA400G3I, SSDSC2BA800G3I to NA36 - SSDSC2BB120G4I, SSDSC2BB240G4I, SSDSC2BB400G4I, SSDSC2BB800G4I to NB37 - fix to prevent possible drive lock up during format.
SATA 1.8"
Initial Code Release - SSDSC1NA200G3I, SSDSC1NA400G3I to NA36 - SSDSC1NB080G4I, SSDSC1NB240G4I, SSDSC1NB400G4I to NB37 - fix to prevent possible drive lock up during format.
=================================================================== IBM Online SAS/SATA Hard Disk Drive Update Program Version 1.14.03-2 Build ID: SAS11403-2
NOTE TO SERVICE - Reference RETAIN TIPS H211424 & H211667 ******************************************* IBM SAS Hard Disk Drive Update Program Version 1.14.03-2 Critical Release Reference RETAIN TIP H211424 & H211667
This package includes updates to drives with product ID strings of:
SAS 3.5"
Initial Code Releases - ST1000NM0023, ST2000NM0023, ST3000NM0023, ST4000NM0023 to BC5B - ST2000NM0043, ST4000NM0043 to EC59 - WD1001FYYG-23S, WD2001FYYG-23S, WD3001FYYG-23S, WD4001FYYG-23S to XA35
Other - ST1000NM0001, ST2000NM0001 to BC4A - support persistent reservation - HUS723020ALS64, HUS723030ALS64 to J3K8 - servo adjustments
SAS 2.5"
Initial Code Releases - ST300MM0006, ST600MM0006, ST900MM0006 to B56H - ST300MM0026, ST600MM0026, ST900MM0026 to E568 - TXA2D20200GA6I, TXA2D20400GA6I, TXA2D20800GA6I, TXA2D21600GA6I to GB76 - see Retain Tip H211424. - HUC101212CSS60 to J2M7 - HUC101212CSS61 to N2M7 - T1200MM0007 to B576 - T1200MM0017 to E574 - ST600MX0001 to BH13 - WD3002BKTG-23E, WD6002BKTG-23E, WD9002BKTG-23E to ZC31 - ST9500431SS to ED12 - ST9146703SS to E536 - ST973352SS to E625 - ST9300553SS to E636 - HUC109030CSS61E, HUC109060CSS61E, HUC109090CSS61E to N2EF - MK3001GRDB, MK6001GRDB, MK9001GRDB to SB35 - drive Que depth fix; see Retain Tip H211667. - HUC109030CSS60, HUC109060CSS60, HUC109090CSS60 to J2EF - HUC109030CSS61E, HUC109060CSS61E, HUC109090CSS61E to N2EF
Other - WD3001BKHG-23D, WD6001BKHG-23D, WD9001BKHG-23D to ZA32 - performance, non-critical fixes and robustness changes.
- ST9600204SS to B54A with maintainance fixes and performance improvements.
- MBD2147RC, MBD2300RC to SB1B - drive Que depth fix; - MBE2073RC, MBE2147RC to SC18 - drive Que depth fix; - MBF2300RC, MBF2600RC to SB2F - drive Que depth fix; - MK1401GRRB, MK3001GRRB to SC2E - drive Que depth fix; see Retain Tip H211667.
- ST9500620SS, ST91000620SS to BD2F - support persistent reservation and maintainance fixes.
SATA 3.5"
Initial Code Release - WD801ABYS-23C0A0, WD1601ABYS-23C0A0 to 06.06H05 - WD5003ABYX-23 to WC38 - added mfg. parameters. - WD2000FYYZ-23UL, WD3000FYYZ-23UL, WD4000FYYZ-23UL to WD36 - ST1000NM0033, ST2000NM0033, ST3000NM0033, ST4000NM0033 to BB58
Other
SATA 2.5"
Initial Code Release - SG9XCS2D50GEIBM to SE6H - WD3000BLHX-23V7BV to YA31 - SSDSC2BA200G3I, SSDSC2BA400G3I, SSDSC2BA800G3I to NA34 - SSDSC2BB120G4I, SSDSC2BB240G4I, SSDSC2BB400G4I, SSDSC2BB800G4I to NB34
Other - ST9250610NS, ST9500620NS, ST91000640NS to BE29 - maintainance updates. - MTFDDAK064MAR-1J1AA, MTFDDAK128MAR-1J1AA, MTFDDAK256MAR-1K1AA, MTFDDAK512MAR-1K1AA to MA57 - added mfg. parameters.
SATA 1.8"
Initial Code Release - SSDSC1NA200G3I, SSDSC1NA400G3I to NA34 - SSDSC1NB080G4I, SSDSC1NB240G4I, SSDSC1NB400G4I to NB34
Other - MTFDDAA064MAR-1J1AB, MTFDDAA128MAR-1J1AB, MTFDDAA256MAR-1K1AB, MTFDDAA512MAR-1K1AB to MA57 - added mfg. parameters.
=================================================================== Version 1.13.01 Build ID: SAS_113
NOTE TO SERVICE - Reference RETAIN H207993
******************************************* IBM SAS Hard Disk Drive Update Program Version 1.13.01 Critical Release Reference RETAIN H207993
This package includes new updates to drives with product ID strings of:
SAS 3.5"
- HUS723020ALS64 to J3K7 Additions to VPD information used by manufacturing and fixes for overall robustness.
- HUS723030ALS64 to J3K7 Initial Code Release
- ST31000424SS, ST32000444SS to BC2D - ST1000NM0001, ST2000NM0001 to BC49 Fixes for overall robustness.
- ST33000650SS to BC3E Initial Code Release
SAS 2.5"
- MK1401GRRB, MK3001GRRB to SC29 Fixes for overall robustness.
- MBF2300RC, MBF2600RC to SB2C Initial Code Release.
- ST9300605SS, ST9600205SS, ST9900805SS to B55D ST9146853SS, ST9300653SS to B63B ST9500620SS, ST91000640SS to BD2E ST9300505SS, ST9600105SS, ST9900705SS to E557 ST9146753SS to E636 Additions to VPD information used by manufacturing and fixes for overall robustness.
SATA 3.5"
- ST500NM0011, ST2000NM0011, ST1000NM0011 to BB4A ST3500514NS, ST31000524NS, ST32000644NS to BB2C Reliability improvements, Additions to VPD information used by manufacturing
- ST33000650NS to BB3A Initial Code Release
- WD5003ABYX-23 to WC36, WD1003FBYX-23Y7B0 to WB35, WD2003FYYS-23W0B0 to WA35 Reliability improvements, Additions to VPD information used by manufacturing
- HUA723020ALA640 to MK7OA910 Additions to VPD information used by manufacturing.
- HUA723030ALA640 to MKAOA910 Initial Code Release
- HUA722050CLA330, HUA722010CLA330 to A3PH Initial Code Release
SATA 2.5"
- TE22D10100GE8IBM to SR3B Initial Code Release
- MTFDDAK064MAR-1J1AA, MTFDDAK128MAR-1J1AA, MTFDDAK256MAR-1K1AA, MTFDDAK512MAR-1K1AA to MA55 Initial Code Release
- ST9250610NS, ST9500620NS, ST91000640NS to BE28 Fixes for overall robustness.
SATA 1.8"
- MTFDDAA064MAR-1J1AB, MTFDDAA128MAR-1J1AB, MTFDDAA256MAR-1K1AB, MTFDDAA512MAR-1K1AB to MA55 Initial Code Release
- TX21B10100GE8IBM, TX21B10200GE8IBM, TX21B10400GE8IBM to SP3A Initial Code Release
Version 1.12.01 Critical Release Reference RETAIN #H207740
This package includes new updates to drives with product ID strings of:
SAS 3.5"
- VPCA300900EST1, VPCA450900EST1, VPCA600900EST1 to A3C4, Corrects firmware bug that may cause data corruption. For further information, please refer to Retain Tip H207740.
- ST3300657SS, ST3450857SS, ST3600057SS to BA5A VPCA300900EST1, VPCA600900EST1 to A3C2 Additions to VPD information used by manufacturing
- ST3750630SS, ST31000640SS to BC1I ST31000424SS, ST32000444SS to BC2C Mode Page change to address performance issues under certain system configurations
- ST33000650SS to BC3D ST1000NM0001, ST2000NM0001 to BC48 Protection Information Fix - see Retain Tip H206140
- HUS723020ALS64, HUS723030ALS64 to J26C Initial Code Release
SAS 2.5"
- ST9300603SS, ST9146803SS to B53E ST9600204SS to B549 Additions to VPD information used by manufacturing
- ST9300605SS, ST9600205SS, ST9900805SS to B55C ST9146853SS, ST9300653SS to B63A ST9500620SS, ST91000640SS to BD2C ST9300505SS to E555 ST9146753SS to E635 Protection Information Fix - see Retain Tip H206140
- ST9600105SS, ST9900705SS to E555 Inital Code Release
- HUC106030CSS60, HUC106060CSS60 to D3A6 Performance improvements, critical fixes for Protection Information (PI) operation - see Retain Tip H206140
- MK1401GRRB, MK3001GRRB to SC27 Additions to VPD information used by manufacturing
- WD9001BKHG-23D, WD6001BKHG-23D, WD3001BKHG-23D to ZA31 Initial Code Release
SATA 3.5"
- WD2502ABYS-23B7A0 to 02.03B08 WD1602ABYS-23B7A0 to 02.03B07 Critical fix to address premature drive failures.
- ST3500514NS, ST31000524NS, ST32000644NS to BB2A ST33000650NS to BB38 ST1000NM0011, ST2000NM0011, ST500NM0011 to BB47 Fixes a potential, but rare data corruption issue which requires a background reallocation that exceeds the error recovery timer.
- HUA723020ALA640 to MK70A900 HUA723030ALA640 to MKA0A900 Initial Code Release
SATA 2.5"
- ST9160511NS, ST9500530NS to BE1A - ST9250610NS, ST9500620NS, ST91000640NS to BE25 Fixes a potential, but rare data corruption issue which requires a background reallocation that exceeds the error recovery timer.
SSD
- TX21B10200GE1IBM, TX21B10400GE1IBM, TX21B10100GE8IBM to PRO6SP36 TE22D10100GE8IBM to PRO6SR36 Initial Code Release
- MTFDDAK128MAR-1J1AA, MTFDDAK256MAR-1K1AA to MA52 Initial Code Release
- MTFDDAA512MAR-1K1AB, MTFDDAK064MAR-1J1AA, MTFDDAK512MAR-1K1AA, MTFDDAA064MAR-1K1AB, MTFDDAA128MAR-1K1AB, MTFDDAA256MAR-1K1AB to MA52 Fixes SMART attribute 190 (Current Temp) which caused warning messages with some versions of Linux
- STEC MACH8 IOPS to 2804 STEC M8PSB2 T4 to SM06 Additions to VPD information used by manufacturing Fixes potential drive failure if power loss occurs during firmware update.
- ST33000650SS to BC3A ST1000NM0001, ST2000NM0001 to BC44 Additions and corrections to VPD information used by manufacturing
SAS 2.5"
- ST9300605SS, ST9600205SS, ST9900805SS to B559 ST9146853SS, ST9300653SS to B637 ST9500620SS, ST91000640SS to BD29 ST9300505SS to E553 SST9146753SS to E633 HUC106030CSS60, HUC106060CSS60 to D390 Additions and corrections to VPD information used by manufacturing
- MK1401GRRB, MK3001GRRB to SC23 Fixes a pottential data corruption bug if there is an unexpected power loss.
SATA 3.5"
- ST3250310NS, ST3500320NS, ST3750330NS, ST31000340NS to BB15 ST3500514NS, ST31000524NS, ST32000644NS to BB29 ST33000650NS to BB36 ST1000NM0011, ST2000NM0011, ST500NM0011 to BB46
SATA SSD
- SG9XCS1B50GMIBM, SG9XCS1F200GEIBM to SA03SB6C SG9XCS2D200GEIBM to SA03SE6C Fix for potential hang during an unexpected power cycle
MTFDDAK128MAR-1JAAB, MTFDDAK256MAR-1JAAB to MA44 Initial Release
This package includes new updates to drives with product ID strings of:
SAS 3.5"
- VPCA300900EST1, VPCA600900EST1 to A3C0 Correction to VPD information used by manufacturing
- ST31000424SS, ST32000444SS to BC2B Fixes to improve robustness from certain read errors due to encroachment
- ST33000650SS to BC37 Fixes to improve overall robustness.
- ST1000NM0001, ST2000NM0001 to BC42 Inital Release
SAS 2.5" (SFF)
- ST9300605SS, ST9600205SS, ST9900805SS to B556 - ST9146853SS, ST9300653SS to B635 Fixes to improve overall robustness.
ST9300505SS to E552 ST9146753SS to E632 Initial Release
- ST9500620SS, ST91000640SS to BD26 Improve general robustness and added support for DPICZ
- HUC106030CSS60, HUC106060CSS60 to D340 Improve general robustness
- MBF2300RC, MBF2600RC to SB28 Improves performance in certain benchmarks
SATA 3.5"
- ST33000650NS to BB34 Fixes to improve overall robustness.
- ST1000NM0011, ST2000NM0011, ST500NM0011 to BB45 Initial Release
- ST3500514NS, ST31000524NS, ST32000644NS to BB29 Fixes for intermittent drive hangs
- ST3250310NS, ST3500320NS, ST3750330NS, ST31000340NS to BB15 Added enhancement for improved robustness
SATA 2.5"
- ST9250610NS, ST9500620NS, ST91000640NS to BE24 Initial Release
- ST9160511NS, ST9500530NS to BE19 Fixes for overall robustness.
SATA 2.5" SSD
- STEC MACH8 IOPS to 2765 STEC M8PSB2 T4 to SM05 Fixes potential drive lock up during cold or warm boots Fixes issues with RHEL 6.x installation
- SG9XCS2D200GEIBM to SA03SE6C Fixes default drive VPD information for manufacturing, the default LED behavior on certain systems and fixes issues with client install thru VMWare 5.x
SATA 1.8" SSD
- STEC M8PSB1 to 0409 STEC M8PSB1 T4 to SM16 Fixes potential drive lock up during cold or warm boots Fixes issues with RHEL 6.x installation
- SG9XCS1B50GMIBM to SA03SB6C SASG9XCS1F200GEIBM to SA03SB6C Fixes potential drive lock up during cold or warm boots and the default LED behavior on certain systems and fixes issues with client install thru VMWare 5.x
This package includes new updates to drives with product ID strings of:
SAS 3.5"
- ST31000424SS, ST32000444SS to BC28 Added enhancement for improved RV robustness
- MBE2073RC, MBE2147RC to SC17 - MBA3073RC, MBA3147RC, MBA3300RC to SA0A Added VPD changes for manufacturing
SAS 2.5" (SFF)
- ST9300603SS, ST9146803SS to B53C ST9600204SS to B548 SST973452SS, ST9146852SS to B62C CBRCA146C3ETS0, CBRCA300C3ETS0 to C610 Added fixes for overall drive robustness
- MBD2147RC, MBD2300RC to SB1A Added VPD changes for manufacturing Addresses issues seen with DSA HDD Diagnostic test.
- MBF2300RC, MBF2600RC to SB25 Addresses issues seen with DSA HDD Diagnostic test.
- ST9500620SS, ST91000640SS to BD22 HUC106030CSS60, HUC106060CSS60 to D2F8 Inital Release
SATA 3.5"
- WD2502ABYS-23B7A0 , WD1602ABYS-23B7A0 to 02.03B07 Addresses firmware issues that may cause a drive to be marked defunct prematurely under certain application workloads.
- ST3250310NS, ST3500320NS, ST3750330NS, ST31000340NS to BB14 ST3500514NS, ST31000524NS, ST32000644NS to BB28 Added enhancement for improved robustness
- WD5003ABYX-23 to WC31 Initial Release
SATA 2.5"
- WD3000BLFS-23YBU0 to 04.04V10 (Critical) Addresses firmware issues that may cause a drive to be marked defunct prematurely under certain application workloads.
- ST9250610NS, ST9500620NS, ST91000640NS to BE21 Initial Release
SATA 2.5" SSD
- STEC MAC8 IOPS to 2732 STEC M8PSB2 T4 to SM04 Addresses an issue installing RHEL 6.0 onto this drive.
SATA 1.8" SSD
- SG9XCS1B50GMIBM, SG9XCS1F200GEIBM to SA03SB63 Addresses rare issue where the drive may not being seen on power up.
- STEC M8PSB1 to 0408 STEC M8PSB1 T4 to SM15 Addresses an issue installing RHEL 6.0 onto this drive.
- ST373455SS, ST3146855SS, ST3300655SS to BA2D - ST373355SS, ST3146755SS, ST3300555SS to BA38 - ST3146356SS, ST3300656SS, ST3450856SS to BA4C - ST3300657SS, ST3450757SS, ST3600057SS to BA58 - ST3750630SS, ST31000640SS to BC1G - ST31000424SS, ST32000444SS to BC26 (All Above Critical) These codes addresses a rare potential for data corruption. For this issue to occur, the drive write cache must be enabled. For further information, please refer to retain tip H197382.
- MBA3073RC, MBA3147RC, MBA3300RC to SA08 This code corrects a firmware bug that may cause the drive to not come ready on a cold boot.
SAS 2.5" (SFF)
- ST973402SS, ST9146802SS to B52B - ST936751SS, ST973451SS to B617 - ST9146803SS, ST9300603SS toB53B - ST9600204SS to B546 - ST973452SS, ST9146852SS to B629 - ST9300503SS to E536 - ST9146752SS to E625 - ST9500430SS to BD14 (All Above Critical) These codes addresses a rare potential for data corruption. For this issue to occur, the drive write cache must be enabled. For further information, please refer to retain tip H197382.
SATA 3.5"
- ST3500514NS, ST31000524NS, ST32000644NS to BB26
- WD2003FYYS-23W0B0 to WA32 - WDC WD1003FBYX-23Y7B0 to WB31 Initial release
SATA 2.5" SSD
- STEC MACH8 IOPS to 2582 (Critical) Corrects firmware bug that may cause data corruption. For further information, please refer to retain tip H197650.
SATA 1.8" SSD
- STEC M8PSB1 to 0405 (Critical) Corrects firmware bug that may cause data corruption. For further information, please refer to retain tip H197650.
This package includes new updates to drives with product ID strings of:
SAS 3.5"
- VPCA300900EST1, VPCA450900EST, VPCA600900EST1 to A3B8 Initial Release
- VPBPA300C3EST1 to A529 Initial Release
- ST3300657SS, ST3450757SS, ST3600057SS to BA56 (Critical) Performance improvements under certain workloads. Corrects a firmware bug that may cause the drive to not come ready on a warm or cold reboot.
- ST31000424SS, ST32000444SS to BC21 Initial Release
SAS 2.5" (SFF)
- ST9300603SS, ST9146803SS to B53A (Critical) Performance improvements under certain workloads. Corrects a firmware bug that may cause the drive to not come ready on a warm or cold reboot.
- ST9600204SS to B544 (Critical) Corrects a firmware bug that may cause the drive to not come ready on a warm or cold reboot.
- ST973452SS, ST9146852SS to B626 Performance improvements under certain workloads.
- ST9500430SS to BD13 Performance improvements under certain workloads
- ST9300503SS to E534 and ST9146752SS to E623 Performance improvements under certain workloads.
- MBD2147RC, MBD2300RC to SB19 (Critical) MBE2073RC, MBE2147RC to SC16 (Critical) Corrects a firmware bug that erroneously reported medium errors which may cause the drive to report PFA alerts and in some cases be marked offline
SATA 3.5"
- WDC WD2502ABYS-23B7A0, WD2502ABYS-23B7A0, WD802ABYS-23B7A0, WD1602ABYS-23B7A0 to 02.03B05 (Critical) Corrects a firmware bug that may cause the drive to not come ready on a warm or cold reboot.
- ST3500514NS, ST31000524NS, ST32000544NS to BB26 (Critical) This code addresses a rare potential for data corruption. For this issue to occur, the drive write cache must be enabled. For further information, please refer to retain tip H197382.
- ST3250310NS, ST3500320NS, ST3750330NS, ST31000340NS to BB13 (Critical) This code addresses a rare potential for data corruption. For this issue to occur, the drive write cache must be enabled. For further information, please refer to retain tip H197382.
- HUA722020ALA330 to JKAOA3GA HUA722050CLA330 to JP2OA3GB HUA722010CLA330 to JP4OA3GB Initial Release
SATA 2.5"
- WD3000BLFS-23YBU0 to 04.04V09 (Critical) This code corrects an additional issue where the drive may be marked defunct.
- ST9160511NS, ST9500530NS to BE18 (Critical) This code addresses a rare potential for data corruption. For this issue to occur, the drive write cache must be enabled. For further information, please refer to retain tip H197382. Also improves performance under certain workloads.
SATA 2.5" SSD
- STEC MACH8 IOPS to 2366 (critical) Corrects firmware bug that may cause degraded performance and in some cases the drive may not come ready after a warm or old boot.
SATA 1.8" SSD
- STEC M8PSB1 to 0404 Turns the default for write cache off Initial Release
IBM SAS Hard Disk Drive Update Program Version 1.06 (Critical)
Added support for the following controllers:
- IBM ServeRAID BR10il SAS controller v2 (49Y4731) - IBM ServeRAID M1015 SAS controller (46M0831) - IBM ServeRAID M5015 SAS controller (46M0829) - IBM ServeRAID M5014 SAS controller (46M0916)
This package includes new updates to drives with product ID strings of:
SAS 3.5"
- ST373455SS, ST3146855SS, ST3300655SS to BA2A ST373355SS, ST3146755SS, ST3300555SS to BA36 (Critical) These codes addresses a rare potential for data corruption. For this issue to occur, the drive write cache must be enabled. When a write is issued while the drive write cache is enabled and a read-verify is performed immediately afterwards, undetected data corruption can occur. For further information, please refer to retain tip H194632.
- ST3750630SS, ST31000640SS to BC1E ST3146356SS, ST3300656SS, ST3450856SS to BA4A This code address an issue where the drive may report an erroneous hardware error to the controller.
- MBA3073RC, MBA3147RC, MBA3300RC to SA07 This code improves sequential write performance under certain workloads when write cache is turned on.
- ST3300657SS, ST3450757SS, ST3600057SS to BA52 NOTE: This is the first code release to be included in the hard drive update package.
SAS 2.5" (SFF)
- ST973402SS, ST9146802SS to B52A ST936751SS, ST973451SS to B616 (Critical) These codes addresses a rare potential for data corruption. For this issue to occur, the drive write cache must be enabled. When a write is issued while the drive write cache is enabled and a read-verify is performed immediately afterwards, undetected data corruption can occur. For further information, please refer to retain tip H194632.
- MBB2073RC, MBB2147RC to SB07 MBD2147RC, MBD2300RC to SB17 MBE2073RC, MBE2147RC to SC14 CBRCA146C3ETS0, CBRCA300C3ETS0 to C370 This code improves sequential write performance under certain workloads when write cache is turned on.
- ST9500430SS to BD11 This is the first code release to be included in the hard drive update package.
SATA 3.5"
- WD2003FYYS-23T8B0 to 01.00D04 This is the first code release to be included in the hard drive update package.
SATA 2.5"
- WD3000BLFS-23YBU0 to 04.04V08 (Critical) This code corrects an issue where the drive may generate a false PFA alert. It also corrects an issue where the drive may be marked defunct
- ST9160511NS, ST9500530NS to BE16 This is the first code releases to be included in the hard drive update package.
=============================================================== IBM SAS Hard Disk Drive Update Program Version 1.05
SAS 2.5" (SFF)
ST9xxxx51SS to B616 ST9xxxx02SS to B52A - This code improves the robustness of write cache operations.
ST9xxxx52SS to B624 ST9xxxx03SS to B536 - This code improves the robustness of write cache operations and sequential write performance when write cache is enabled
MBD2xxxRC to SB14 - Initial release
SAS 3.5"
ST3xxxxx6SS to BA49 ST3750630SS to BC1D - This code improves the robustness of write cache operations.
SATA 3.5"
WDC WD2502ABYSxxxxx tp 02.03B04 - Fixes incorrect VPD data
SATA 2.5"
WD3000xxxxx to 04.04V05 - Adjusted PHY setting for better signal quality margin.
SATA 2.5" SSD
STEC MACH8 IOPS to 2269 GBRL0xxxxx to 4.17 GBRLBxxxxx to 4.26 - The new firmware will correct a rare data error condition in the drives where invalid data may be passed to the system without any error condition flagged.
Due to the severity of this potential issue, IBM strongly recommends applying this firmware as soon as possible.
For further information, please refer to retain tip H195663. =================================================================
IBM SAS Hard Disk Drive Update Program Version 1.04
SAS 2.5" (SFF)
CBRBAxxxxxx to C49B - Modified VPD for JSxx Blades
MBBxxxxxx to SB05 - Fixed a performance issue when synchronizing or rebuilding an array when used in a LSI 1064 based SAS controller
MBC2073RC to SC04 - Fixed a performance issue when synchronizing or rebuilding an array when used in a LSI 1064 based SAS controller
SAS 3.5"
MBAxxxxxx to SA06 - Fixed a performance issue when synchronizing or rebuilding an array when used in a LSI 1064 based SAS controller
ST3xxxxx6SS to BA48 - This code addresses a rare potential for data corruption. For this issue to occur, the drive write cache must be enabled. When a write is issued while the drive write cache is enabled and a read-verify is performed immediately afterwards, undetected data corruption can occur. For further information, please refer to retain tip H194632
ST3750630SS to BC1A - This code addresses a rare potential for data corruption. For this issue to occur, the drive write cache must be enabled. When a write is issued while the drive write cache is enabled and a read-verify is performed immediately afterwards, undetected data corruption can occur. For further information, please refer to retain tip H194632
- Fixes a hang condition during failover operation on a Bladecenter-S chassis with an RSSM controller.
BBAxxxxxx to BH0K - Fixes an issue where a drive may prematurely report a PFA trip
GNAxxxxxx to BH0J - Fixes an issue where a drive may prematurely report a PFA trip
SATA 3.5"
ST31000340NS, ST370330NS, ST3250310NS, ST3500320NS to BB12 - This code addresses a rare potential for data corruption. For this issue to occur, the drive write cache must be enabled. When a write is issued while the drive write cache is enabled and a read-verify is performed immediately afterwards, undetected data corruption can occur. For further information, please refer to retain tip H194632
- This code addresses a rare potential for data loss. After a power cycle, the drive will no longer be available and becomes unresponsive. Data will become inaccessible due to the drive not responding. For further information, please refer to retain tip H194623.
WDC WD2502ABYSxxxxx tp 02.03B04 - Fixes incorrect VPD data
SATA 2.5"
WD3000xxxxx to 04.04V04 - Fixes a Critical issue - IBM recommends a backup of the data on the drive before attempting to update the firmware. This is a fix for an issue where data required for spinup of the drive may be overwritten rendering the drive inoperable on a subsequent warm or cold system reboot.
- Fixes a issue with an internal timer that may cause drive to go offline when used in a non-raid environment.
IBM SAS Hard Disk Drive Update Program Version 1.03
Added support for the following controllers:
IBM ServeRAID MR10il SAS Controller IBM ServeRAID MR10is SAS Controller IBM ServeRAID BR10i SAS Controller IBM ServeRAID BR10il SAS/SATA Controller IBM SAS/SATA RAID Kit IBM SAS Expansion Card (CFFv) for IBM BladeCenter IBM SAS Connectivity Card (CFFv) for IBM BladeCenter
New updates to drives with product ID strings of:
SAS 2.5" (SFF)
ST9146802SS, ST973402SS to B529 - Added support for JS2x Blades in a Bladecenter-S chassis - Picked up the latest network roll of firmware fixes to address some unexpected system hangs,
ST936751SS, ST973451SS to B615 - Added support for JS2x Blades in a Bladecenter-S chassis - Picked up the latest network roll of firmware fixes to address some unexpected system hangs,
CBRBAxxxxxx to C495 - Initial release
MBBxxxxxx to SB04 - Initial release
MBC2073RC to SC03 - Initial release
SAS 3.5"
VPBAxxxxxx to A496 - Added support for JS2x Blades in a Bladecenter-S chassis
ST3146855SS, ST3300655SS, ST373455SS to BA29 - Added support for JS2x Blades in a Bladecenter-S chassis - Picked up the latest network roll of firmware fixes to address some unexpected system hangs,
ST3450856SS to BA47 - IBM was notified by our vendor that there is a defect in a level of firmware used in the IBM 450GB 15K SAS 3.5" HS HDD (PN 41Y8454) in option 42D0519, which under certain conditions may lead to data being written or read back incorrectly on the drive. IBM has worked closely with the vendor to identify the problem and developed a firmware fix to resolve the problem. Customers should upgrade to firmware level BA47
MBAxxxxxx to SA05 - Initial release
ST3750630SS to BC17 - Initial release
SATA 3.5"
ST3250620NS, ST3500630NS, ST3750640NS to 3.AEV
ST3250621NS, ST3500631NS, ST3750641NS to 4AEV
ST31000340NS, ST3750330NS, ST3250310NS, ST3500320NS to BB10 - Initial release
HUA721050KLA330, HUA721075KLA330, HUA721010KLA330, to AB4A - Initial Release
SATA 2.5" SSD
GBRL001xxx, GBRL0031xxx to 4.14 - corrects intermittent hangs during operations =============================================================
IBM SAS Hard Disk Drive Update Program Version 1.02
Support of ServeRaid 8s Raid controller Support of ServeRaid 10i Raid controller Support of ServeRaid 10k Raid controller Support of ServeRaid 10m Raid controller Support of LSI 1078 SAS controller Support of SATA drives attached to SAS controllers
Support and updates to drives with product ID strings of:
VPAxxxxxx to A650 Change to improve performance in sequential write workloads Improvements to internal drive logging and media checking. Critical Fix: This code addresses a rare potential for data loss This code update should be applied to prevent the event from occurring.
ST340XXXAS and ST380XXXAS to 3.48 Fix to address Read Log Command abort error.
ST340XXXAS, ST380XXXAS and ST3160XXXAS to 3.05 Fix to address intermittent hard drive no-shows after system comes out of hibernation.
ST3500641NS to 3.AES Changes to the refresh rate of the DRAM used on the drive.
ST3500640NS to 3.AEJ Initial release.
HDTXX... & HDSXX... codes go to Masked ROM version. HDS725050KLA360 to AQ1A to also Fix an issue that occurs while running in NCQ mode with write cache disabled.
ST3xxxx4SS to BA26 Improves intialization times with some ServeRaid controllers Performance improvements under some sequential workloads.
ST3xxxx5SS to BA35 Improves intialization times with some ServeRaid controllers
ST9xxx51SS to B612 Intial release of firmware
ST9xxxx2SS to B526 Improves intialization times with some ServeRaid controllers
MAUxxxx and MAXxxxx Added support for new SAS features required by new SAS controllers.
IBM SAS Hard Disk Drive Update Program Version 1.01 Please refer to RETAIN tip H186019.
Support of ServeRaid 8e Raid controller Support of ServeRaid 8k Raid controller Support of ServeRaid 8l Raid controller Support of on-board LSI 1064 controller IBM MegaRaid 8480 Raid Controller
Firmware updates to drives with Product ID strings:
GNAxxxx and BBAxxxx to BH0G Picks up the latest network roll of firmware fixes to improve performance under some sequential workloads and improvements to signal quality parameters.
ST9xxxx to B51D amd ST3xxxx to BA18 Picks the latest network roll of firmware fixes to address unexpected system hangs and some performance improvements.
MAVxxxx to S116 and MAYxxxx to T107 Picks up the latest network roll of firmware fixes and to improve signal quality parameters.
MAXxxxx to T904 Initial release ============================================================
IBM SAS Hard Disk Drive Update Program Version 1.00 Initial Release
Support of Adaptec 48xxx based controllers Support of ServeRaid 8i Raid controller Support and updates to drives with product ID strings of: ST936701SS ST973401SS ST336754SS ST373454SS GNAxxxxx BBAXXXXX
ibm_fw_hdd_sas-1_14_04-1_windows_32-64_chg
Integrated Management Module II (IMM2) Firmware Update Change History
Version 4.21, 1AOO58U - Updated the firmware with patches to address the OpenSSL vulnerabilities listed in CVE-2014-0224, CVE-2014-0221, CVE-2014-0195, CVE-2014-3470, CVE-2010-5298, CVE-2014-0198
Version 4.20, 1AOO58R - Fixed an issue in I2C communications that could cause CMM to report, "FPGA Host 2 Communication is offline" - Fixed a problem that required the browser cache to be cleared before accessing IMM2 web page. - Corrected a problem that caused the CMM to reporting an I/O connectivity problem with the FSM - Enhanced the IMM2 service and support data to better aid in the diagnosis of VPD issues - Added support for the S3 system sleep state - Added support for agentless monitoring of PCIe devices - Enabled a Flex compute node to be powered on sooner after installation in the chassis - Added RFC compliance checking for user input - Ensured consistency in Daylight Savings Time setting between the CMM and IMM2 - Added support for enhanced hard drive monitoring and status - Corrected an issue that caused false reports of incorrect / missing / corrupt VPD events.
Version 3.79, 1AOO56G - Critical Fix for x240, x222, x440, x220, Flex System Manager Node -- Updated the firmware with patches to address the 'Heartbleed' OpenSSL bug (CVE-2014-0160)
Version 3.77, 1AOO56H - Critical Fix for x3630 M4, x3530 M4 -- Updated the firmware with patches to address the 'Heartbleed' OpenSSL bug (CVE-2014-0160)
Version 3.75, 1AOO56F - Critical Fix for x3650 M4 BD -- Updated the firmware with patches to address the 'Heartbleed' OpenSSL bug (CVE-2014-0160)
Version 3.73, 1AOO56D - Critical Fix for dx360 M4, x3650 M4, x3550 M4, x3500 M4, x3100 M4, x3250 M4, x3750 M4 -- Updated the firmware with patches to address the 'Heartbleed' OpenSSL bug (CVE-2014-0160)
Version 3.71, 1AOO52W - Critical Fix for x3550 M4: -- Fixed a potential security vulnerability with information contained in service data (refer to security bulletin CVE-2014-0882) -- Fixed a problem system may shut down with new Mezz card installed
Version 3.70, 1AOO52Q - Critical Fix: -- Fixed a potential security vulnerability with information contained in service data (refer to security bulletin CVE-2014-0882)
Version 3.67, 1AOO50G - Supports System x3650 M4 BD, Machine Type: 5466
Version 3.65, 1AOO50D NOTE: If you have a System x3550 M4 with the "Emulex VFA Mezz card" and a ==== single processor, don't update the IMM2 with this firmware - Critical Fix: -- Fixed a problem on the System x3650 M4 server where the system shuts down due to a false temperature event from the ambient temperature sensor - Fixed a problem where the IMM2 may erroneously report a fan fault in the x3630 M4 when the system is powering on. See RETAIN tip H207461 for the description of the original problem - Fixed a screen refresh problem where it occurs after resetting IMM2 - Fixed a firmware problem that may cause the IMM2 to reset unexpectedly - Fixed a problem where the IMM2 SNMP interface may intermittently be unavailable if the system is running VMWARE ESXi 5.x - Fixed a problem where the CPU may potentially throttle in the x3650 M4 system with two 550 watt power supplies and one hard drive backplane
Version 3.60, 1AOO50C - Added support for the Intel IvyBridge-EP CPU on the HS23 blade - Fixed a problem where the HS23 hard drive number reported in the AMM event did not match the drive number on the physical blade enclosure - Fixed an intermittent problem where the UXSPI update utility may erroneously report that the firmware image is not meant for the HS23 system - Fixed a problem that causes the AMM web interface to not display properly when drives are removed or inserted - Fixed a problem where the IMM2 may not detect an over temperature condition for a DIMM
Version 3.56, 1AOO50K - Fixed a Flex Systems Manager (FSM) Configuration Patterns problem where configuration deployments would fail with an activation job error message: ERROR: LDAP failed with rc=102 ERROR: LDAP failure: SYSPROF_LDAP_BIND_FAIL INFO
Version 3.55, 1AOO50E - Fixed an issue so a fully qualified domain name can be used for a syslog collector in addition to an IP address - Fixed an issue to allow a user the ability to specify the port to use for syslog notifications, rather than only supporting the default port
Version 3.50, 1AOO50B - Fixed an issue where the default static IPv4 address was assigned to the IMM2 management interface instead of the user configured static IPv4 address in DHCP fallback to static mode, if the DHCP server was not available - Prevent a VLAN from being configured on the IMM2 management interface for Flex nodes - Fixed a problem where the CPU throttling state recovery indication to the FSM might not be properly reported - Corrected an issue with the launch of the IMM2 Remote Console operation that required a restart of the IMM2 - Corrected the firmware to apply the inactivity timeout to web pages that automatically refresh - Updated the web browser support. The following web browsers are supported: Firefox 3.6-20 Chrome 13-26 Internet Explorer 8-10 - Added support for the S3 Sleeping State to the IMM2 user interfaces on those servers that include the S3 feature - Added changes to use the minimum TLS version setting that is provisioned by the CMM in Flex for encrypted communication - Enhanced the content of the service data file to assist in problem determination - Added a option to have the IMM2 automatically copy the firmware from its primary bank into the backup bank after the firmware has been run for a period of time and proven itself to be stable - Enhanced the IMM2 firmware to automatically restore itself without the need for user intervention if the firmware becomes corrupted - Added new commands on the command line interface to stop a serial redirection operation that is currently in progress, and to report the user names that have a serial redirection operation started - Added changes to use the DNS settings that are provisioned by the CMM in Flex to support use of an LDAP hostname for authentication - Updated the firmware to log hardware inventory change informational messages - Added support for the Service Advisor feature on stand alone rack servers to automatically call for service when a repair is needed - Added support to configure the Ethernet-over-USB IP address using ASU - Added changes to support the thermal requirements for new Solid State Drives used with the ServeRAID M5100 Series IBM Flex System Flash Kit v2 - Updated the IMM2 date and time web page on Flex nodes to only display the current date and time - Updated the SSH firmware to address various OpenSSH vulnerabilities that have been identified by security community - Added changes to support the Web Services-Management protocol - Corrected a problem that caused the IMM2 to reboot after running continuously for several months
Version 3.40, 1AOO48M - Supports System x3250 M5, Machine Type: 5458
Version 3.37, 1AOO48P - Added support for the IBM NeXtScale Storage Native Expansion (NeX) Tray for IBM NeXtScale nx360 M4
Version 3.36, 1AOO48Q - Fixed a problem on System x iDataPlex dx360M4 where a CIM client cannot get needed event information for asserted events
Version 3.35, 1AOO48N - Critical Fix: -- Corrected a problem on the System x3650 M4 and x3650 M4 HD servers causing the system to shut down due to an erroneous critical temperature event reported by the ambient temperature sensor
Version 3.30, 1AOO48L - Fixed a problem on System x iDataPlex dx360M4 where nodes shutdown due to a critical temperature event from the Ambient Temp sensor
Version 3.20, 1AOO48K - Supports System NeXtScale nx360 M4, Machine Type: 5455
Version 3.10, 1AOO48H - Added support for System x3650 M4 HD, Machine Type 5460 - Added support for newer CPUs in the System x iDataPlex dx360M4, x3500M4, x3550M4, and x3650M4 servers - Added a feature on the System x3550 M4 and x3650 M4 to fail over from the dedicated systems management network port to the shared port when the dedicated network link is not available - Fixed a problem where an NTP update message is logged every 1440 minutes - Fixed a problem where the IMM2 web interface won't accept an NTP hostname string that starts with a numeric character - Fixed a problem where the LDAP Group Filter authentication does not work as expected - Fixed a problem where the IMM2 reports 0 Watts when the System x iDataPlex dx360M4 is DC cycled or resumes from the S3 sleep state - Fixed an Ethernet problem that may cause an IMM2 failure to establish a link when the Ethernet speed and duplex settings are not configured to use auto-negotiation
Version 2.60, 1AOO42Y NOTE: For increased security, only high strength ciphers are now supported when using https.
==== Your browser must support one of the following cipher suites: | HE-RSA-AES256-SHA256 TLSv1.2 Kx=DH Au=RSA Enc=AES(256) Mac=SHA256 DHE-RSA-AES256-SHA SSLv3 Kx=DH Au=RSA Enc=AES(256) Mac=SHA1 AES256-SHA256 TLSv1.2 Kx=RSA Au=RSA Enc=AES(256) Mac=SHA256 AES256-SHA SSLv3 Kx=RSA Au=RSA Enc=AES(256) Mac=SHA1 DHE-RSA-AES128-SHA25 TLSv1.2 Kx=DH Au=RSA Enc=AES(128) Mac=SHA256 DHE-RSA-AES128-SHA SSLv3 Kx=DH Au=RSA Enc=AES(128) Mac=SHA1 AES128-SHA256 TLSv1.2 Kx=RSA Au=RSA Enc=AES(128) Mac=SHA256 AES128-SHA SSLv3 Kx=RSA Au=RSA Enc=AES(128) Mac=SHA1
- Added support for the IBM Flex System x222 Compute Node - Improved Remote Presence (keyboard, mouse and virtual media)and in-band management (LAN-over-USB) performance - Changed the default file name for the Service Data file (First Failure Data Capture or FFDC) on the web interface to incorporate the machine type, serial number and a timestamp - Updated the IMM2 PCIe Bridge Firmware to support Windows 2012 Certification - Improved IMM2 Reliability, Availability, Serviceability (RAS) Features: - Added component and system VPD integrity checking & reporting - Enhanced Service Data collection (FFDC)
- Enhanced the configuration of UEFI settings via the CIM interface for applications like IBM Systems Director - Added Sub-manufacturer inventory support for mezz card and Sub-manufacturer inventory support for the system for IBM Systems Director - Added the UUID for the PCIe Expansion Node and the Storage Expansion Node to the SLP attributes reported by the IMM2 - Added the number of enabled CPU core and the Static power savings mode support for IBM Systems Director - Enhanced the alert recipient email notification, firmware update, event log, log out, and hardware information web pages to improve usability - Added support for IPMI & SNMPv3 accounts to be centrally managed by the Flex Chassis Management Module - Fixed a problem where an invalid configuration was erroneously reported for an unpopulated CPU socket in the IBM Flex System x440 - Enabled users to configure a longer trespass message - Increased the length of the LDAP group filter field to 511 characters - Corrected a problem with ServeRAID M5115 SAS/SATA Controller for IBM Flex Systems energy pack monitoring - Corrected a problem with ServeRAID M5115 SAS/SATA Controller for IBM Flex Systems LED management - Added IPMI command to configure the LAN-over-USB IP address
Version 2.52, 1AOO42E - Critical Fix: -- Fixed a problem on the System x3300 M4 servers causing the system to shut down due to a critical temperature event reported by the ambient temperature sensor
Version 2.50, 1AOO40Z - Added support for the Solarflare SFN5162F MR Dual Port 10GbE SFP+ Adapter, and Solarflare SFN6122F LL Dual Port 10GbE SFP+ Adapter for System x3550 M4 and x3650 M4 NOTE: UEFI also needs to be upgraded to D7E128B (v 1.40 or newer, for the x3550 M4) or VVE128B (v1.40 or newer, for the x3650 M4) to support these cards - Added support for the Mellanox ConnectX-3 Dual Port QDR/FDR10 mezzanine card and Mellanox ConnectX-3 VPI Single-port QSFsorry.P FDR14/40GbE HCA on System x3550 M4, and x3650 M4 - Added optional Fan Zone support for single CPU systems (x3550 M4 and x3650 M4) with an installed with Qlogic Dual Port 10GbE SFP+ Embedded VFA - Fixed an intermittent problem where a Gratuitous ARP is not sent successfully - Fixed a problem where after the IMM2 firmware is flashed, there is a slight chance that the IMM2 web server can't be accessed when AC power is reapplied - Disabled the weak cipher suit used in the SSL/TLS firmware with the https protocol to avoid potential security vulnerabilities - Fixed a problem on the System x3750 M4 where the IMM2 shared mode network connection will be lost after upgrading the firmware of the onboard Emulex BE3 controller - Fixed a problem on the HS23 where the OS watchdog timer doesn't work - Fixed a problem on the HS23 where the Fault LED occasionally lights when the chassis is powered on with the BPE4 - Fixed a problem on the HS23 where the server may shut down after resetting the IMM2 - Enhanced the messaging mechanism between the BladeCenter AMM and the IMM2 - Added power management capability in the IMM2 web interface
Version 2.06, 1AOO42B - Corrected the firmware to prevent users from disabling the IPv6 interface - Fixed a problem that caused the IMM2 to reset itself after running for a period of time - Corrected the SNMP request to disable the IMM2 IPv4 interface so that it disables only IPV4, instead of both IPv4 and IPv6 - Fixed a problem that occasionally caused the server to shut down when the IMM2 is reset - Fixed a problem where the IMM2 incorrectly reported a battery voltage error - Corrected a problem where repeated frequent IMM2 resets causes the IMM2 to run the firmware in its backup partition - Corrected a problem report where the IMM2 did not report a failure to the system profiles script when the chassis UUID is unavailable
Version 2.00, 1AOO40E - Critical Fix: -- Improves the reliability of the CPU and/or memory VRDs. This improvement lowers the occurrence of CPU & memory VRD failures on the x3300 M4 and x3750 M4 -- IMM2 improved error reporting for missing/down CPU in invalid CPU configuration for x4440 M4 - Corrected a problem that caused 3.3V and 5V faults to be reported by IMM2 firmware level 1aoo34z(1.85) on the HS23E blade - Eagle Ray SAS controller LED fault support improvements added for all ITE platforms
Version 1.97, 1AOO36R - Critical Fix: -- Improves the reliability of the CPU and/or memory VRDs. This improvement lowers the occurrence of CPU & memory VRD failures on x3650 M4, x3550 M4 and x3500 M4 - Thermal behavior enhancement on System x iDataPlex dx360 M4
Version 1.95, 1AOO36P (for System x iDataPlex servers) - Added support for NVIDIA K1,K2,K20,K20x GPUs on System x iDataPlex dx360 M4
Version 1.88, 1AOO40A (for PureFlex System Compute Nodes) - Critical Fix: -- Fixed a potential security vulnerability with node management user accounts (Reference to Security Bulletin CVE-2012-4838) - Fixed a problem on Flex Systems where the SNMP get and set authority levels did not work - Fixed a problem on Flex Systems where mounting of remote media in a Linux environment showed a USB error - Fixed a problem on Flex Systems that caused the Remote Control Video Viewer window on a RHEL5.8 client to move when the arrow key is pressed - Fixed a problem on Flex Systems where the SMTP "Test Connection" button did not work when the SMTP port is changed - Fixed a problem on Flex Systems where the SNMP get and set authority levels did not work - Fixed a problem on Flex Systems where mounting of remote media in a Linux environment showed a USB error - Fixed a problem on Flex Systems that caused the Remote Control Video Viewer window on a RHEL5.8 client to move when the arrow key is pressed - Fixed a problem on Flex Systems where the SMTP "Test Connection" button did not work when the SMTP port is changed - Corrected the Mezzanine card power fault monitoring on the Flex System x440 Compute Node - Corrected a problem with IBM Fabric Manager configuration from the Flex Chassis Management Module - Added support to retrieve configuration settings from an LDAP server - Added support for the Flex Storage Expansion Node - Added support for CIM indications over https - Added fingerprint checking for SFTP transfers - Added automatic recovery of corrupted UEFI firmware - Added support for authenticated SMTP alerts - Added a display of the users who have a Remote Control session in progress - Provided a request to a Remote Control client to disconnect the session - Added support for the server S3 (sleep) state - Changed the default NTP sync interval from 3 minutes to 24 hours - Fixed a problem where the ambient temperature is only displayed in degrees Fahrenheit - Fixed a problem on the web interface where 'local' is not accepted as a top level domain name in the email format when configuring event recipients - Corrected an IMM2 SNMP trap OID to match the prior IMM format - Fixed a problem that prevented an external USB device from being remotely mounted as a floppy drive in Windows Server 2003 - Changed the blue Location led behavior to not blink when a firmware update is in progress
Version 1.85, 1AOO34Z (for BladeCenter blades and System x Servers) - Critical Fix: -- Fixed a potential security vulnerability with node management user accounts (Reference to Security Bulletin CVE-2012-4838) - Fixed a problem where a Blade Open Fabric Manager uncorrectable error can occur when booting the HS23 - Fixed a problem that IMM IPMI command periodically timed out when using HS23 in Chassis with the management network being disabled in some special configurations - Added support for the HS23 to recognize NEBS/Enterprise Mode for Telco chassis (this feature needs Advanced Management Module Firmware BPET62U or later) - Fixed a problem where the ambient temperature is only displayed in degrees Fahrenheit - Fixed a problem where on rack systems that have 2 power supplies installed, the 'Redundancy Power unit has been restored' event is logged every time that the IMM2 is reset - Fixed a problem where a Fan Tach going low (lower critical) event is intermittently asserted when configuring the IMM2 to use an NTP server - Fixed a problem on the web interface where 'local' is not accepted as a top level domain name in the email format when configuring event recipients - Corrected an IMM2 SNMP trap OID to match the prior IMM format - Fixed a problem that prevented an external USB device from being remotely mounted as a floppy drive in Windows Server 2003 - Changed the blue Location led behavior to not blink when a firmware update is in progress - Fixed a problem on the HS23E that causes Preboot DSA to hang in CLI mode when a user accesses it using Serial over LAN - Fixed a problem on rack servers where ASU cannot set IMM.GRP_GroupName.1 and IMM.GRP_GroupName.2 if the string length is 16 characters or longer. The firmware was corrected to allow a maximum length of 64 characters - Added support for CIM indications over https - Added fingerprint checking for SFTP transfers - Added automatic recovery of corrupted UEFI firmware - Added support for authenticated SMTP alerts - Added a display of the users who have a Remote Control session in progress - Provided a request to a Remote Control client to disconnect the session - Added support for the server S3 (sleep) state
Version 1.85, 1AOO34Y (for PureFlex System V7000 Storage Node) - Added support for Flex System V7000 Storage Node
Version 1.79, 1AOO36F - Thermal behavior enhancement on System x iDataPlex dx360 M4
Version 1.78, 1AOO36E - Thermal behavior enhancement for 115W & 95W CPU on System x iDataPlex dx360 M4 - Added support for the Intel KNC 5110p card
Version 1.77, 1AOO32X - Thermal behavior enhancement for 115W CPU on System x iDataPlex dx360 M4
Version 1.75, 1AOO32S - Fixed a problem that a PCI uncorrectable error occurs when using HS23 with CFFh card (44W4479) in the BCS chassis - Fixed a problem where an "Unsupported blade/blower combination" error is reported when using the HS23 in the BCHT chassis configured in Enterprise mode - Fixed a problem in System x3530 M4, where the reported Active Energy Manager input/output power readings are inaccurate - Fixed a problem in System x3530 M4, x3630 M4, where an LDAP account can not login using the AD_CrossDomain tree - Fixed a problem in System x3530 M4, x3630 M4, where a user cannot use the "DNS to Find LDAP server/Extract search domain from login ID" setting from the IMM2 web to access an eDirectory server - Added support for Common Form Factor power supplies Oversubscription in System x3750 M4, Machine Types: 8722, 8733
Version 1.65, 1AOO32O - Support the power supply Over-Subscription feature - Support the 750 Watt DC power supply for System x3650 M4,x3550 M4 - Support IBM 1.2TB High IOPS MLC Mono Adapter, IBM 2.4TB High IOPS MLC Duo Adapter for System x3650 M4, x3550 M4 - Support IBM LLM-SM Dual Port 10GbE SFP+ Adapter for System x3650 M4 - Fixed an abnormal Fan Zone recovery problem - Fixed a problem that IMM2 Email notification doesn't support an e-mail address with the underscore '_' character - Fixed a power policy "Restore" problem - Fixed a problem where the LDAP hostname can't be changed by ASU - Fixed a problem that the length of email address (for SNMP notification) is limited to 32 characters
Version 1.60, 1AOO32P - Fixed a problem where the Chassis Management Module cannot correctly get the Mezzanine card information for the Flex System x220 Compute Node - Fixed a problem where the PreBoot DSA firmware could not be updated under VMWare ESXi 5 through the Flex System Manager's Update Manager - Fixed a problem where an IMM2 communications error can occur if an unexpected response causes VMware to spawn multiple processes when it does not receive the data it expects from the IMM2
Version 1.59, 1AOO32K - Supports System x3300 M4, Machine Types: 7382
Version 1.53, 1AOO30Z - Included an update to the FPGA that fixes an intermittent issue where the CPUs do not show up in a system with a RAID card installed, when the AC is removed and reapplied from the server without allowing sufficient time for the system hardware to completely discharge
Version 1.52, 1AOO34A - Added redundant power supply and hot swap hard drive support in System x3100 M4
Version 1.51, 1AOO30Y - Support for NVIDIA Tesla K10, M2070Q, M2090 GPU cards on System x iDataPlex dx360 M4 - Fixed a problem in System x iDataPlex dx360 M4 where the password expires after creating a new user in the IMM2 web interface - Fixed a problem in System x iDataPlex dx360 M4, when the memory sensor changes from 'critical' to 'warning,' but the severity is not updated in the syslog or the email notification
Version 1.50, 1AOO30W - Supports System HS23, Machine Types: 7875,1929 - Supports System x3650 M4, Machine Types: 7915 - Supports System x3550 M4, Machine Types: 7914 - Supports System x3500 M4, Machine Types: 7383 - Added support for the Intel Xeon E5-2643 processor on the HS23 in BCH-R3, BCS and BCHT chassis - Fixed a problem where the IMM2 incorrectly logged an "Unsupported blade/blower combination" on the HS23 in the BC-E 8677 chassis - Fixed a power restore problem on systems that have 2 Delta power supplies installed - Fixed a problem in System x3650 M4, x3550 M4, x3500 M4, where OpenLDAP Group Authentication filtering function doesn't work - Fixed a problem in System x3650 M4, x3550 M4, x3500 M4, where LDAP account can not login using the AD_CrossDomain tree
Version 1.45, 1AOO28S - Fixed a problem in Flex System x220 where the Chassis Management Module (CMM) Command Line Interface can't display the Mezzanine card information - Fixed a problem in Flex System x220 and x240 where the prebootDSA firmware cannot be updated through Flex System Manager's Update Manager
Version 1.40, 1AOO30T - Supports System x3750 M4, Machine Types: 8722, 8733
Version 1.38, 1AOO30D - Supports System x3630 M4, Machine Type: 7158 - Supports System x3530 M4, Machine Type: 7160
Version 1.36, 1AOO30P - Fixed an intermittent remote media failure in system HS23E - Fixed a problem in system HS23E where remote media does not work when two or more users have concurrent remote control sessions
Version 1.34, 1AOO28Q - Supports Flex System Manager Node, Machine Type: 8731 - Supports Flex System x240 Compute Node, Machine Types: 7863, 8737, 8738 - Supports Flex System x220 Compute Node, Machine Types: 2585, 7906
Version 1.32, 1AOO28N - Supports System x3630 M4, Machine Type: 7158 - Supports System x3530 M4, Machine Type: 7160 - Fixed a problem that may see unexpected fan events after power on in x3100 M4 and x3250 M4 systems - Enhanced robustness on I2C bus test
Version 1.30, 1AOO28M - Supports System HS23E, Machine Types: 8038,8039
Version 1.26, 1AOO26N - Supports System x3650 M4, Machine Types: 7915
Version 1.25, 1AOO26K - Supports System x3550 M4, Machine Types: 7914 - Supports System x3500 M4, Machine Types: 7383 - Supports System x iDataPlex dx360 M4, Machine Types: 7912,7913
Version 1.22, 1AOO26O - Supports System HS23, Machine Types: 7875,1929 Problems Fixed: - Fixed a problem where the BOFM configuration may be lost after a blade power cycle - Fixed an LED error indication that occurs on the BPE4 - Fixed a problem where the Info LED lights during the discovery stage - Fixed a problem where an ABR event is not logged - Fixed an intermittent communication problem between the AMM and IMM2 - Fixed a problem that causes the remote media function to hang when a second user attempts to mount remote media - Fixed a problem where the 'break' keyboard function doesn't work with Serial Over LAN
Version 1.21, 1AOO26L - Supports System HS23, Machine Types: 7875,1929
Version 1.10, 1AOO24B - Supports System dx360 M4, Machine Types: 7918,7919
Version 1.05, 1AOO10K - Fixed the false alert issue in redundant PSU model
Version 1.00, 1AOO10I - Initial Release - Supports System x3250 M4, Machine Types: 2583,2587 - Supports System x3100 M4, Machine Types: 2582,2586
(C) Copyright International Business Machines Corporation 2008 - 2014. All rights reserved. ibm_fw_imm2_1aoo64l-4_50_anyos_noarch_chg
Intel® Chipset Device Software
Production Version 9.4.0.1026
Release Notes
August 8, 2013
Microsoft Windows* 2000
Microsoft Windows XP*
Microsoft Windows XP* Professional x64 Edition
Microsoft Windows Server* 2003
Microsoft Windows Server* 2003 x64 Edition
Microsoft Windows Vista* RTM and SP’s
Microsoft Windows Server* 2008 R2
Microsoft Windows Server* 2008 R2 x64 Edition
Microsoft Windows* 7
Microsoft Windows* 7 x64
Microsoft Windows* 8
Microsoft Windows* 8 x64
Microsoft Windows Server 2012
Microsoft Windows Server 2012 x64
Microsoft Windows* 8.1
Microsoft Windows* 8.1 x64
Microsoft Windows Server 2012 R2
Microsoft Windows Server 2012 R2 x64
INFORMATION IN THIS DOCUMENT IS PROVIDED IN CONNECTION WITH INTEL PRODUCTS. NO LICENSE, EXPRESS OR IMPLIED, BY ESTOPPEL OR OTHERWISE, TO ANY INTELLECTUAL PROPERTY RIGHTS IS GRANTED BY THIS DOCUMENT. EXCEPT AS PROVIDED IN INTEL'S TERMS AND CONDITIONS OF SALE FOR SUCH PRODUCTS, INTEL ASSUMES NO LIABILITY WHATSOEVER AND INTEL DISCLAIMS ANY EXPRESS OR IMPLIED WARRANTY, RELATING TO SALE AND/OR USE OF INTEL PRODUCTS INCLUDING LIABILITY OR WARRANTIES RELATING TO FITNESS FOR A PARTICULAR PURPOSE, MERCHANTABILITY, OR INFRINGEMENT OF ANY PATENT, COPYRIGHT OR OTHER INTELLECTUAL PROPERTY RIGHT.
A "Mission Critical Application" is any application in which failure of the Intel Product could result, directly or indirectly, in personal injury or death. SHOULD YOU PURCHASE OR USE INTEL'S PRODUCTS FOR ANY SUCH MISSION CRITICAL APPLICATION, YOU SHALL INDEMNIFY AND HOLD INTEL AND ITS SUBSIDIARIES, SUBCONTRACTORS AND AFFILIATES, AND THE DIRECTORS, OFFICERS, AND EMPLOYEES OF EACH, HARMLESS AGAINST ALL CLAIMS COSTS, DAMAGES, AND EXPENSES AND REASONABLE ATTORNEYS' FEES ARISING OUT OF, DIRECTLY OR INDIRECTLY, ANY CLAIM OF PRODUCT LIABILITY, PERSONAL INJURY, OR DEATH ARISING IN ANY WAY OUT OF SUCH MISSION CRITICAL APPLICATION, WHETHER OR NOT INTEL OR ITS SUBCONTRACTOR WAS NEGLIGENT IN THE DESIGN, MANUFACTURE, OR WARNING OF THE INTEL PRODUCT OR ANY OF ITS PARTS.
Intel may make changes to specifications and product descriptions at any time, without notice. Designers must not rely on the absence or characteristics of any features or instructions marked "reserved" or "undefined". Intel reserves these for future definition and shall have no responsibility whatsoever for conflicts or incompatibilities arising from future changes to them. The information here is subject to change without notice. Do not finalize a design with this information.
The products described in this document may contain design defects or errors known as errata which may cause the product to deviate from published specifications. Current characterized errata are available on request.
Contact your local Intel sales office or your distributor to obtain the latest specifications and before placing your product order.
Code names featured are used internally within Intel to identify products that are in development and not yet publicly announced for release. Customers, licensees and other third parties are not authorized by Intel to use code names in advertising, promotion
The Intel® Chipset Device Software may contain design defects or errors described as “known issues” which may cause the product to deviate from expected performance. Current characterized known issues are documented in this update.
Intel and the Intel logo are trademarks of Intel Corporation in the U.S. and other countries.
*Other names and brands may be claimed as the property of others.
The Intel® Chipset Device Software contains support for the following:
· Intel® 8 Series Chipset Family
· Intel® 7 Series Express Chipset
· Intel® 6 Series Express Chipsets
· Intel® 5 Series Chipsets
· Intel® 4 Series Chipsets
· Intel® 3400 Series Chipsets
· Intel® 3 Series Chipsets
· Intel® 900 Series Chipsets
· Intel® 800 Series Chipsets
· Intel® Server/Workstation Chipsets
· Intel® Xeon® Processor 5500/3500 series
· 4th Generation Intel® Core™ processor family
· 3rd Generation Intel® Core™ processor family
· 2nd Generation Intel® Core™ processor family
New Features
Features Added in Baseline 9.4.0.1026
No New feature added.
Features Added in Baseline 9.4.0.1022
ID
Description
1
Certified for Windows 8.1 support.
Added Intel Sensor Solution HWIDs duplicate for VEN_INT&DEV_33D1.
2
Added Intel Sensor Solution HWIDs duplicate for VEN_INT&DEV_33D1.
3
Added support for MUP specification 2.4.1.
Features Added in Baseline 9.4.0.1017
ID
Description
1
Updated Branding for devices 0c08, 8C56, 8C52, 8C54
Features Added in Baseline Production Version 9.4.0.1016
ID
Description
1
Fixed missing cat file lxptusb.cat, Ivytown.cat
Features Added in Baseline Production Version 9.4.0.1014
ID
Description
1
Updated Installer with new INFs for extract driver switch
2
Added device information for Intel sensor solution for Intel® 8 Series chipset family (previously code named Lynx Point-LP)
3
Fixed readme.txt
Features Added in Baseline Beta 9.4.0.1012
ID
Description
1
Added Crystal Well DID
2
Updated Branding for Intel® 8 Series chipset family devices (previously code named Lynx Point-LP)
Features Added in Baseline Beta 9.4.0.1010
ID
Description
1
Updated Readme with Windows 8 and Windows Server 2012 instructions
2
Updated all 4th Generation Intel® Core™ processor family and Intel® 8 Series chipset family devices with correct branding
3
Signed with HCK
Features Added in Baseline Alpha 9.4.0.1007
ID
Description
1
Updated Intel® 8 Series chipset family (previously code named Lynx Point-LP) Device IDs
Features Added in Baseline Alpha 9.4.0.1003
ID
Description
1
Updated 4th Generation Intel® Core™ processor family Devices
2
Added Intel Corporation Signing to all installation on Windows 8
Known Issues
Known Issues in Build 1026
ID
Description
1
When running the Chipset INF installer on Windows 8, Windows 8.1, Windows Server 2012, or Windows Server 2012 R2, there is a delay before any UI is displayed. This delay is usually approximately 2 minutes, but can be much more.
Known Issues in Build 1022
ID
Description
1
When running the Chipset INF installer on Windows 8, Windows 8.1, Windows Server 2012, or Windows Server 2012 R2, there is a delay before any UI is displayed. This delay is usually approximately 2 minutes, but can be much more.
Known Issues in Build 1017
ID
Description
1
Centerton files are 0 bytes
Issues Resolved
Issues Resolved in Build 1026
ID
Description
1
Fixed USB INF files contain errors
2
Various string fixes
3
Updated "PDH" to "PCH" in readme.
4
Updated section names
Issues Resolved in Build 1022
ID
Description
1
Fixed missing newline after branding string.
2
Fixed Chipset INF files not signed issue.
3
Fixed device 33D1 display string.
Issues Resolved in Build 1017
ID
Description
1
Fixed AHCI bug filed in MSFT Database [REG:113011610144431] Intel CCG: BSOD 0x7B after uninstalling Intel RST software.
2
Fixed Sensor Co-Installer issue (Sensor hub is not working because of incorrectly installed by Intel chipset software)
Issues Resolved in Build 1016
ID
Description
1
Fixed the error message “This computer currently contains software version … which is newer than the version you are about to install”
Issues Resolved in Build 1011
ID
Description
1
Fixes the error message “This computer does not meet the minimum requirement “ on double install/update chipset software
Issues Resolved in Build 1003
ID
Description
1
None
intc_dd_inteldp_9_4_0_1026_windows_32-64_chg
Dynamic System Analysis (DSA) Preboot Change History
Version 9.30, Build ID DSYTB27 o New System Support: Flex System x440 Compute Node (7917, 2584) IBM System x3300 M4 (7382) o Problem(s) Fixed: 1. PMR - 05766,YL2,760 Aenservice appeared after running UpdateXpress & DSA. 2. PMR - 05193,W01,000 7147 - Intermittent random PSOD in vmware. multiple
o Enhancement: None o Limitations: - FIFO test of NIC diagnostics is not supported on the Intel I350 Quad Port GbE. Powerville has shared FIFO and shared FIFO registers making a test impossible if four ports are running at once. - For more limitations please refer to readme. o Special usage: - 2012 ToolsCenter 12D
Version 9.29, Build ID DSYTAB6 o New System Support: Flex System x440 Compute Node (7917, 2584) o Problem(s) Fixed: None o Enhancement: None o Limitations: - DSA implemented signature in product build. Please update IMM to 1AOO34V or later before updating pDSA, or flash failure will occur; - The user can not flash Embedded Preboot on VMware ESX3.5 prior to U4. - When flashing DSA Preboot using wflash/lflash, if you notice a build mismatch error, please make sure IMM firmware level is minimum 29B. Once you upgrade the IMM firmware to 29B or higher, DSA Preboot will continue to flash without error; - For more limitations please refer to readme. o Special usage: - NGP-4Q(12B+ FixPack3)
Version 9.28, Build ID DSYTAA4 o New System Support: x3630 M4 (7519) x3550 M4 (7383) o Problem(s) Fixed: None o Enhancement: None o Limitations: - DSA implemented signature in product build. Please update IMM to 1AOO34V or later before updating pDSA, or flash failure will occur; - The user can not flash Embedded Preboot on VMware ESX3.5 prior to U4. - When flashing DSA Preboot using wflash/lflash, if you notice a build mismatch error, please make sure IMM firmware level is minimum 29B. Once you upgrade the IMM firmware to 29B or higher, DSA Preboot will continue to flash without error; - For more limitations please refer to readme. o Special usage: - IMMV2 system 2012 4Q-LFC
Version 9.27, Build ID DSYTA7F o New System Support: None o Problem(s) Fixed: None o Enhancement: None o Limitations: Refer to readme o Special usage: - IMMv1 system 2012 4Q-LFC
Version 9.26, Build ID DSYTA6G o New System Support: None o Problem(s) Fixed: None o Enhancement: DSA Performance Optimization o Limitations: Refer to readme o Special usage: - 2012 ToolsCenter 12C
Version 9.21, Build ID DSYTA1U o New System Support: None o Problem(s) Fixed: 1.Recovered LSI Rawdata: "MegaCli -FwTermLog -Dsply -aALL" o Enhancement: None o Limitations: Refer to readme o Special usage: - 2012 ToolsCenter 12B Plus
Version 9.20, Build ID DSYTA0R o New System Support: x3530 M4 (7160) x3630 M4 (7158) x3750 M4 (8722, 8733) HS23E (8038, 8039) Flex System x220 Compute Node (7906, 2585) o Problem(s) Fixed: 1. PMR 05819,YL2,760 An error occured during the execution of DSA on Windows, DSA exits with an error at the eventlog plugin. o Enhancement: 1. Add HW remote presence key collection for High volume systems. o Limitations: Refer to readme o Special usage: - 2012 ToolsCenter 12B
Version 9.07, Build ID DSYT9CA o New System Support: x3750 M4 (8722, 8733) o Problem(s) Fixed: None o Enhancement: None o Limitations: None o Special usage: None
Version 9.06, Build ID DSYT9BC o New System Support: Flex System x220 Compute Node (7906) o Problem(s) Fixed: None o Enhancement: None o Limitations: None o Special usage: None
Version 9.05, Build ID DSYT9AA o New System Support: x3630 M4 (7158) x3530 M4 (7160) x3100 M4 Refresh1 (2582) x3250 M4 Refresh1 (2583) o Problem(s) Fixed: None o Enhancement: None o Limitations: None o Special usage: None
Version 9.04, Build ID DSYT99E o New System Support: Flex System x240 Compute Node (8737, 8738, 7863) o Problem(s) Fixed: None o Enhancement: None o Limitations: None o Special usage: None
Version 9.03, Build ID DSYT98I o New System Support: HS23E (8038, 8039) o Problem(s) Fixed: None o Enhancement: None o Limitations: None o Special usage: None
Version 9.02, Build ID DSYT97C o New System Support: iDataPlex dx360 M4 server/2U chassis (7912/7913) o Problem(s) Fixed: None o Enhancement: None o Limitations: None o Special usage: None
Version 9.01, Build ID DSYT96D o New System Support: x3500 M4 (7383) x3550 M4 (7914) x3650 M4 (7915) o Problem(s) Fixed: o Enhancement: None o Limitations: None o Special usage: None
Version 9.00, Build ID DSYT95W o New System Support: BladeCenter HS23 (7875) BladeCenter HX5 (7873, 1910) Flex System x240 Compute Node (8737, 8738, 7863) Flex System Manager Node (8735) x3500 M4 (7383) x3550 M4 (7914) x3630 M4 (7158) x3650 M4 (7915) x3690 X5 (7192, 7147) x3850 X5 / x3950 X5 (7143, 7191) o Problem(s) Fixed: 1. PMR 24817,8QG,000, PMR 30680,499,000 After kicking off DSA on one of the management nodes (x3650 M2) of SONAS 1.2(RHEL5 based), the node became unresponsive and dropped out of the cluster.
2. PMR 53575,000,834, PMR 05961,YL4,760 After running DSA, when using linux dump command on a DDS5 usb internal tape drive attached in x3650M2 backup speed becomes very slowly or failed.
3. PMR 26905,095,866 Running DSA 4.0 as well as DSA 3.4 on x3850 X5 the server is freezing, users have to reboot the server.
4. DSA 4.00 and previous version segmentation fault issue
6. PMR 15086,OTH,760 DSA 9.00 with --no-imm-lan option does not work correctly on RHEL5.3. The USB port state is changed during the execution of DSA.
o Enhancement: 1. Add GUI and CLI menu for Features on Demand (FoD) Activation Key Management o Limitations: Refer to readme. o Special usage: None
Version 4.02, Build ID DSYT92I o New System Support: iDataPlex Direct Water Cooled dx360 M4 server o Problem(s) Fixed: None o Enhancement: None o Limitations: Refer to readme. o Special usage: None
Version 4.01, Build ID DSYT92H o New System Support: None o Problem(s) Fixed: None o Enhancement: None o Limitations: Refer to readme. o Special usage: - 2011 4Q lifecycle
Version 4.00, Build ID DSYT91S o New System Support: None o New server options support: None o Problem(s) Fixed: 1. PMR 15065,OTR,760 On Unisys ES7000 or x3850 X5, the Memory Card number with failed DIMM can not be reported in IPMI Eventlog if the Memory Card is bigger than 4.
o Enhancement: 1. Added > 1TB memory test support o Limitations: Refer to readme. o Special usage: - 2011 ToolsCenter 11D
Version 3.48, Build ID DSYT89PUS o New System Support: None o Problem(s) Fixed: None o Enhancement: None o Limitations: Refer to readme. o Special usage: - 2011 3Q lifecycle
Version 3.47, Build ID DSYT89LUS o New System Support: x3100 M4 (2582) x3250 M4 (2583) o Problem(s) Fixed: None o Enhancement: None o Limitations: Refer to readme. o Special usage: None
Version 3.46, Build ID DSYT89GUS o New System Support: System x iDataPlex dx360 M3 (6391) o Problem(s) Fixed: None o Enhancement: None o Limitations: Refer to readme. o Special usage: None
Version 3.45, Build ID DSYT89EUS o New System Support: None o Problem(s) Fixed: None o Enhancement: None o Limitations: Refer to readme. o Special usage: - x3850 X5 / x3950 X5 (7143, 7191) EXA scaling
Version 3.43, Build ID DSYT87HUS o New System Support: BladeCenter HX5 (7873,1910) o Problem(s) Fixed: None o Enhancement: None o Limitations: Refer to readme. o Special usage: None
Version 3.43, Build ID DSYT87FUS o New System Support: x3850 X5 / x3950 X5 (7143, 7191) x3690 X5 (7192, 7147) o Problem(s) Fixed: o Enhancement: None o Limitations: Refer to readme. o Special usage: None
Version 3.40, Build ID DSYT85T o New System Support: BladeCenter HS22 (1911) x3650 M3 (5454) o Problem(s) Fixed: None o Enhancement: None o Limitations: Refer to readme. o Special usage: None
Version 3.30, Build ID DSYT80S o New System Support: None o Problem(s) Fixed: None o Enhancement: 1. Correct a problem with LSI MegaRAID reporting for IBM Service o Limitations: Refer to readme. o Special usage: None
Version 3.20, Build ID DSYT75X o New System Support: x3630 M3 (7377) x3690 X5 (7148, 7149) o Problem(s) Fixed: None o Enhancement: 1. Add LSI Hard Disk Drive (HDD) diagnostic back for both the onboard SAS chip and ServeRAID o Limitations: - If you have ServeRAID (M1015) SAS/SATA Controller, 46M0831 installed, you will need one of the following: - uEFI GYE133A or greater for x3200 M3 and x3250 M3 - uEFI Y4E145B or greater for x3400 M2, x3400 M3, x3500 M2 and x3500 M3 - uEFI D6E145C or greater for x3550 M2, x3550 M3, x3650 M2 and x3650 M3 - For more limitations please refer to readme o Special usage: None
Verion 3.20, Build ID DSYT75U o New System Support: x3755 M3 (7164) o Problem(s) Fixed: None o Enhancement: None o Limitations: None o Special usage: None
Verion 3.10, Build ID DSYT70X o New System Support: x3400 M3(7378, 7379) x3500 M3(7380) x3550 M3 (7944, 4254) x3620 M3 (7376) x3650 M3 (7945, 4255) x3690 X5 (7148, 7149) x3850 X5 Standard Single-node and Multi-node models(7145, 7146) x3950 X5 Specially configured Database Engine models (7145, 7146) iDataplex dx360 M3 (6391) HS22V (7871,1949) HX5 (7872, 1909) o Problem(s) Fixed: None o Enhancement: 1. Added ServeRAID data collection o Limitations: - The following function has been temporarily removed from DSA Preboot for both the onboard SAS chip and ServeRAID: - LSI Hard Disk Drive (HDD) diagnostic - If you have ServeRAID (M1015) SAS/SATA Controller, 46M0831 installed, you will need one of the following: - uEFI GYE133A or greater for x3200 M3 and x3250 M3 - uEFI Y4E145B or greater for x3400 M2, x3400 M3, x3500 M2 and x3500 M3 - uEFI D6E145C or greater for x3550 M2, x3550 M3, x3650 M2 and x3650 M3 - For more limitations please refer to readme o Special usage: None
Version 3.02.56, Build ID DSYT56e o New System Support: None o Problem(s) Fixed: None o Enhancement: - The following systems are supported when BoMC downloads the corresponding system enablement pack: HS22V (7871,1949) x3550 M3 (7944,4254) x3650 M3 (7945,4255) x3400 M3 (7378,7379) x3500 M3 (7839) dx360 M3 (6391) o Limitations: - No HDD diagnostics available - To ensure quality and stability of the DSA code, we have reverted some display functionality of RAID information to what was shown in previous versions of DSA. This affects RAID display on the following adapters: Megaraid 8480 Serveraid MR10i Serveraid MR10is Serveraid MR10m Serveraid MR10k Serveraid M1015 Serveraid M5014 Serveraid M5015 On these adapters, the RAID information is generated from the output of separate command line tools and the format may not match other output in DSA. - For more limitations please refer to readme o Special usage: None
Version 3.00, Build ID: DSYT55a o New System Support: None o Problem(s) Fixed: None o Enhancement: - Added support for Machine Types 7327, 7328, 4251, 4252, 4261 o Limitations: - No HDD diagnostics available - For more limitations please refer to readme o Special usage: None
Version 1.16, Y4YT19AUS o New System Support: None o Problem(s) Fixed: None o Enhancement: - Added support for Machine Types 7839, 7836, 7837 o Limitations: None o Special usage: None
Version 1.15, TMYT20AUS o New System Support: None o Problem(s) Fixed: None o Enhancement: - Added support for Machine Types 7321, 6380, 7323 o Limitations: None o Special usage: None
Version 1.14, D6YT37AUS o New System Support: None o Problem(s) Fixed: None o Enhancement: - Added support for Machine Types 7946, 4198, 7947, 4199 o Limitations: None o Special usage: None
Version 1.13, P9YT48AUS o New System Support: None o Problem(s) Fixed: None o Enhancement: - Added support for Machine Types 7870, 1936, 7809 o Limitations: None o Special usage: None
Version 1.00 - Initial release
(C) Copyright International Business Machines Corporation 2008 - 2012. All rights reserved. ibm_fw_dsa_dsyte2f-9_61_anyos_anycpu_chg
=============================================================================== Emulex 16GB FC Firmware Package ===============================================================================
Firmware Version: 10.2.261.36 Supported On: IBM System x and Flex
Problems Fixed: - Server will halt on black screen during rolling reboot of Windows 2012 x64 uEFI SANboot - Server intermittently fails to boot during rolling SANboot (any OS) - Fixed a boot hang associated with server BIOS change
Incremental Interoperability: - Inventory support for the Flex xHMC (Chassis Hardware Management Console)
Known Issues: - To disable BFS after BOFM or IFM have been enabled, it must be done by setting BFS manually in the EFIBoot config utility or sending a BOFM default command to the port. This is working as expected and we are documenting it here.(159504)
Firmware Version: 1.1.43.21 Supported On: IBM System x
Incremental Interoperability: - None
Problems Fixed: - Server will halt on black screen during UEFI rolling reboot reboot cycle test with 16Gb HBA under Windows 2012 x64 - Allows installing and booting from an MSA2040 array configured in PT-PT direct mode at 16G speed. We're now able to enable Autoscan and install/boot from multiple OSes - Added support to bypass configuring pnp Boot Table in a multipath situation. Fixes Microsoft Windows Server 2008 bug. - Fixed the "Read Rev Failure" issue during POST to accept "0" address value as a valid. - Fixed the wrong VPD HBA model number display on the banner and properties, and change the proper VPD data field size. - Fixed the timeout issue on Flogi when try to select a HBA device without connecting to a target or FC switch. - System will halt during reboot cycle test under Windows 2012 x64 UEFI - Fixing PCI Function Number decoding issue in BOFM code - Boot fails during SLES11 SP3 rolling reboot - Enable BFS and update the NVRAM if BOFM target is available - Add Diagnostic support for UEFI 16Gb FC - Reconnect menu item causes a System BIOS exit request only on AMI core system BIOS - Adding port's protocol name and WWPN in browser help menu
Firmware Version: 1.1.43.12 Supported On: IBM System x and Flex
Incremental Interoperability: - IBM Flex System FC5052 2-port 16Gb FC Adapter - IBM Flex System FC5054 4-port 16Gb FC Adapter - FC support for support for ASU/Config Patterns for Flex and ASU for IBM System x & BladeCenter - Direct PT - PT connections - Extended SCSI Pass Thru Protocol - IBM Tape unit interoperability - see SSIC for more information - SLES 11.3 on IBM System x
Problems Fixed: - Windows 2012 OS will not resume from Sleep mode - Boot Device entries can be lost after exiting from BIOS setup
Firmware Version: 1.1.21.9 Supported On: IBM System x
Incremental Interoperability: - IBM Server machine types. Refer to IBM ServerProven for specifics
Problems Fixed: - Fixed FCP_CONF handling bug - No SRR sent after drop write data frame, I/O fails - Adapter does not invoke recovery after dropped read data frame - I/O throughput performance improvements - Link failover when connected to FC Tape - Cosmetic: Cleaned up miscellaneous error handling messages in debug log files - If a new device comes online and attempts to acquire an AL_PA after the link has already come up, the FC link appears to hang until the cable is pulled - Cleaned up link negotiation at 4GB/8GB/16GB speeds with specific switches
uEFI Flash Update Version 1.00 (BuildID N2E108N) Change History
Version 1.00 - BuildID: N2E108N ------------------------------------------------- Problem(s) Fixed: Fixed an issue leading to a 3x boot failure after uefi update Fixed an issue in which system could hang after creating RAID 10 Fixed an issue causing Rhel 6.5 installation to hang
Enhancements: Added 8-socket (4 node) support Added support for legacy Windows 2012 installs Limitations: Dependencies:
Version 1.00 - BuildID: N2E108M (Initial Release) ------------------------------------------------- Problem(s) Fixed:
Enhancements:
Limitations:
Dependencies: Use with IMM level 58R or later ibm_fw_uefi_n2e108n-1_00_anyos_32-64_chg
=============================================== Change History =============================================== Document Name: Change History Document Status: IBM General Avalibility Document Version: 3.2.4.1 Latest Driver Version: 3.2.4.1 Document Updated: 04/14/2014 Product: Brocade FC/NIC/CNA
=================================================
Please visit www.brocade.com/hba for full documentation, additional drivers, and management applications.
Version 3.2.4.1 Fixes and Enhancements: - AIM and UCM fix
Version 3.2.3.0 New Features and Feature Enhancement: - Support for patching firmware via boot image - Support for UEFI Health Check Protocol
Fixes and Enhancements: - Enhanced Windows Network performance
Version 3.2.1.1 Fixes and Enhancements: - Firmware update to address the thermal notification issue
Version 3.2.1.0 New Features and Feature Enhancement: - Support for "IBM Flex System FC5024D 4-port 16Gb FC Adapter" PART_NUMBER "95Y2381"
Fixes and Enhancements: - Enhanced windows Hyper-V driver to avoid a infrequent crash seen during shutdown with VMQ enabled
Version 3.2.0.0 Release for FC/NIC/CNA
New Features and Feature Enhancement: - Updated code for Brocade 1867 IBM Flex System FC5022 FC Adapter for x22 - BB Credit Recovery - Forward Error Correction (FEC) - Diagnostic Port (D-Port) - LUN Level IO Latency/Stats - UCM (Unified Configuration Manager) - Windows Network Performance and scalability enhancement
Fixes or enhancements: - For Brocade 8x5/4x5, I/O interrupt latency values are dynamically updated. Changes are effective for all the ports on the given adapter. - Addressed infrequent PSOD in VMware ESX with iSCSI or VLAN configured traffic in DCB enabled network. - Enhanced BR1860 Tape IO performance. - Improved BR1860 initialization on Xen kernel.
Version 3.1.0.1 Release for 16G FC/NIC/CNA
Adapter Features - Updated code for Brocade 1867 IBM 16G Mezzanine Fabric Adapter.
HBA - Fibre Channel 16G Support
Fixes and Enhancements: N/A
Version 3.0.3.1 Initial release for 16G FC/NIC/CNA
Adapter Features - Initial support for Brocade 1867 IBM 16G Mezzanine Fabric Adapter.
HBA - Fibre Channel 16G Support
Fixes and Enhancements: N/A brcd_fw_cna_3_2_4_1_a_windows_32-64_chg
Firmware Version: 10.2.261.36 Supported On: IBM System x, BladeCenter, and Flex
Problems Fixed: - FRU VPD fields are now properly populated on CN4054 and CN4054R - Fixed UE resulting slow boot at UEFI splash screen and NIC devices removed - When using VMQ with Windows Server 2012 or 2012R2, the user may experience VM connectivity loss, packet drops, system hangs, inability to shutdown VMs and possible system crashes on shutdown. The complete solution also requires 10.2 based Windows NIC driver. - Resolved an issue for auto negotiation between the x240 onboard LoM to the EN2092 1Gb switch
Incremental Interoperability: - Inventory support for the Flex xHMC (Chassis Hardware Management Console)
Known Issues: - Server hangs at "Scan for Fibre Devices", when port disable/enable performed on the Brocade Converged 10 Gb Ethernet Switch Module for IBM BladeCenter. Do not force the link down after POST has completed and then attempt a discovery of the SAN. If you do a reboot is needed.(159125)
- To disable BFS after BOFM or IFM have been enabled, it must be done by setting BFS manually in the EFIBoot config utility or sending a BOFM default command to the port. This is working as expected and we are documenting it here.(159505)
- Intermittently, 255 vPorts can not be created successfully in Windows with the error "ERROR: <127>: Not enough HBA or fabric resources. VPort not created!". There is no known workaround other than to reduce the number of configured vPorts.(155807)
- Target login will fail if you dynamically reconfigure the SAN after POST has executed. If you do, then reboot the system.(159121)
- PXESelect showing incorrect MAC address, bandwidth & SVID values. This bug is a cosmetic issue and only impacts PXESelect. The values which are not displayed correctly in PXESelect can be viewed in UEFI, OCM, and on the switch.(151140)
- IFM Boot target is consumed on a particular port even though the target is not accessible to that port. IFM/BOFM boot target configuration is not supported. This limitation has existed for several code releases and we are now documenting it here.(159226)
- "Configurations have changed" message seen in UEFI setup screen when no changes were made. This pop-up can be ignored and will be fixed in a future code release.(159086)
- The server may hang at "Connecting Boot Devices and Adapters" if both FCoE ports have boot targets configured when using a 10Gb Ethernet Pass-Thru Module for IBM BladeCenter. The server will not hang if only a single port is configured with a boot target.(157137)
- FCoE VLAN ID is being displayed incorrectly as hexadecimal in UEFI menu. This is a cosmetic issue only. The correct VLAN is properly assigned, so there is no functional impact. The actual VLAN ID can be observed elsewhere. (159561)
- In FCoE configurations with BE3 in UFP mode on System x3690 X5 servers, a UEFI rescan is required to boot to the OS. This renders UFP not supportable in UFP/FCoE configurations(158669)
- After making changes to any options in the sub-menus under 'controller configuration' in the Emulex UEFI NIC configuration utility and escaping to the prior menu, there is no prompt to save the changes. Be sure to select 'Save' before exiting the configuration menu. Applies to some legacy rack and blade servers.
- Intermittently, when changing the multichannel mode, the system will not auto-reboot after exiting F1 UEFI setup and will continue through POST. Manually reboot the server to ensure all changes are applied. Applies to some legacy rack and blade servers.
- When using a Cisco Nexus 4001i Switch Module, intermittently the 10Gb link may come up as 1Gb link speed. An internal switch port bounce will typically establish a 10Gb link. Setting the switch port speed to fixed 10000 will also allow a guaranteed 10Gb connection. More info available in the Cisco Nexus 4001i Switch Module firmware release notes.
Firmware Version: 10.0.803.13 Supported On: IBM System x
Problems Fixed: - IBM Feature on Demand for storage keys now works with Optimized Boot enabled - Link detection time improved from ~70 secs on DAC cable insertion/removal
Incremental Interoperability: - Unified Fabric Protocol on IBM System x Servers
Known Issues: - After making changes to any options in the sub-menus under 'controller configuration' in the Emulex UEFI NIC configuration utility and escaping to the prior menu, there is no prompt to save the changes. Be sure to select 'Save' before exiting the configuration menu. Applies to some legacy rack and blade servers.
- Intermittently, when changing the multichannel mode, the system will not auto-reboot after exiting F1 UEFI setup and will continue through POST. Manually reboot the server to ensure all changes are applied. Applies to some legacy rack and blade servers.
- When using a Cisco Nexus 4001i Switch Module, intermittently the 10Gb link may come up as 1Gb link speed. An internal switch port bounce will typically establish a 10Gb link. Setting the switch port speed to fixed 10000 will also allow a guaranteed 10Gb connection. More info available in the Cisco Nexus 4001i Switch Module firmware release notes.
Firmware version: 4.6.281.26 (supported on IBM System x, BladeCenter and Flex)
Problems Fixed: - CN4054 VPD is not valid during/after IMM flash - Links go down after 497 to 498 days in VMWare - Send 'link down' to switch if UE occurs on ASIC
Incremental Interoperability: - Emulex Dual Port 10GbE SFP+ VFA IIIr for IBM System x - Emulex Dual Port 10GbE SFP+ Embedded VFA IIIr for IBM System x
Known Issues: - After making changes to any options in the sub-menus under 'controller configuration' in the Emulex UEFI NIC configuration utility and escaping to the prior menu, there is no prompt to save the changes. Be sure to select 'Save' before exiting the configuration menu. Applies to some legacy rack and blade servers.
- Intermittently, when changing the multichannel mode, the system will not auto-reboot after exiting F1 UEFI setup and will continue through POST. Manually reboot the server to ensure all changes are applied. Applies to some legacy rack and blade servers.
- When using a Cisco Nexus 4001i Switch Module, intermittently the 10Gb link may come up as 1Gb link speed. An internal switch port bounce will typically establish a 10Gb link. Setting the switch port speed to fixed 10000 will also allow a guaranteed 10Gb connection. More info available in the Cisco Nexus 4001i Switch Module firmware release notes.
Firmware version: 4.6.281.21 (supported on IBM System x, BladeCenter and Flex)
Problems Fixed: - OCe10 VFA CNAs stuck in "Switch Independent Mode" - Port Identification fails on Port 1 under uEFI menu for PCIe rack VFA
Incremental Interoperability: - IBM Flex System CN4054R 10Gb Virtual Fabric Adapter - IBM Ivy Bridge CPU Based Servers
Known Issues: - After making changes to any options in the sub-menus under 'controller configuration' in the Emulex UEFI NIC configuration utility and escaping to the prior menu, there is no prompt to save the changes. Be sure to select 'Save' before exiting the configuration menu. Applies to some legacy rack and blade servers.
- Intermittently, when changing the multichannel mode, the system will not auto-reboot after exiting F1 UEFI setup and will continue through POST. Manually reboot the server to ensure all changes are applied. Applies to some legacy rack and blade servers.
- When using a Cisco Nexus 4001i Switch Module, intermittently the 10Gb link may come up as 1Gb link speed. An internal switch port bounce will typically establish a 10Gb link. Setting the switch port speed to fixed 10000 will also allow a guaranteed 10Gb connection. More info available in the Cisco Nexus 4001i Switch Module firmware release notes.
Firmware version: 4.6.281.8 Supported On: IBM System x, BladeCenter, and Flex
Problems Fixed: - HW iSCSI offload is now supported in UFP Mode (UFP limited to IBM Flex) - Server may hang during boot with multichannel disabled with switch default settings. The IBM 10Gb Virtual Fabric Switch has trunking enabled and the Brocade Converged 10Gbe Switch Module has ports in "Access" mode. If the Brocade Converged 10Gbe Switch Module ports are switched to trunking mode, the hang no longer occurs. - When using Config Patterns with virtual MACs the bandwidth settings are are intermittently reset after some number of server reboots.
Incremental Interoperability: - IBM Flex System Fabric SI4093 System Interconnect Module - IBM Flex System EN6131 40Gb Ethernet Switch - IBM Flex System x222 10Gb Virtual Fabric LOM - SLES 11.3 on IBM System x and BladeCenter - Complete NIC/iSCSI/FCoE support for ASU/Config Patterns for Flex and ASU for IBM System x & BladeCenter
Known Issues: - After making changes to any options in the sub-menus under 'controller configuration' in the Emulex UEFI NIC configuration utility and escaping to the prior menu, there is no prompt to save the changes. Be sure to select 'Save' before exiting the configuration menu. Applies to some legacy rack and blade servers. - Intermittently, when changing the multichannel mode, the system will not auto-reboot after exiting F1 UEFI setup and will continue through POST. Manually reboot the server to ensure all changes are applied. Applies to some legacy rack and blade servers. - When using a Cisco Nexus 4001i Switch Module, intermittently the 10Gb link may come up as 1Gb link speed. An internal switch port bounce will typically establish a 10Gb link. Setting the switch port speed to fixed 10000 will also allow a guaranteed 10Gb connection. More info available in the Cisco Nexus 4001i Switch Module firmware release notes.
Firmware version: 4.6.166.7 Supported On: IBM System x and BladeCenter
Problems Fixed: - When connected to Cisco Nexus 4k and 5k switches, the virtual link may not recover on a switch port bounce (shut/no shut) in an ESXi environment. A switch port re-bounce will recover the link. - When connected to Cisco Nexus 4k and 5k switches, OneCommand Manager reports CRC's and there is poor network performance in an ESXi environment
Incremental Interoperability: - ASU/Config Pattern configuration for iSCSI and FCoE (BE3) - UFP for iSCSI (BE3) - Support for using 10Gb ports with legacy 1Gb switches installed in a BladeCenter-S chassis or MSIM/MSIM-T expansion units in a BladeCenter-H or -HT chassis - Update Express System Packs now support Linux errata kernels - ESXi 5.1
Known Issues: - After making changes to any options in the sub-menus under 'controller configuration' in the Emulex UEFI NIC configuration utility and escaping to the prior menu, there is no prompt to save the changes. Be sure to select 'Save' before exiting the configuration menu. Applies to some legacy rack and blade servers. - Intermittently, when changing the multichannel mode, the system will not auto-reboot after exiting F1 UEFI setup and will continue through POST. Manually reboot the server to ensure all changes are applied. Applies to some legacy rack and blade servers. - Server may hang during boot with multichannel disabled with switch default settings. The IBM 10Gb Virtual Fabric Switch has trunking enabled and the Brocade Converged 10Gbe Switch Module has ports in "Access" mode. If the Brocade Converged 10Gbe Switch Module ports are switched to trunking mode, the hang no longer occurs. - When using a Cisco Nexus 4001i Switch Module, intermittently the 10Gb link may come up as 1Gb link speed. An internal switch port bounce will typically establish a 10Gb link. Setting the switch port speed to fixed 10000 will also allow a guaranteed 10Gb connection. More info available in the Cisco Nexus 4001i Switch Module firmware release notes.
Firmware Version: 4.4.180.3 Supported On: IBM Flex
Problems Fixed: - 10Gb ports intermittenly linking at 1Gb speed with the IBM Flex System Fabric EN4093/EN4093R 10Gb Scalable Switch. - Switch Configuration after an erase configuration is not reflected after reboot - 'Logical Admin Link' not enabled after applying BOFM in Switch Agnostic mode - IBM pre-boot Diagnostics NIC physical loopback test fails in Command and GUI mode - Multiple cosmetic issues with iSCSI UEFI BIOS - 10Gb ports do not detect 'Link Down' with some switches - When adding an iSCSI hardware target in UEFI through discovery login, the boot target setting is ignored.
Interoperability since previous release: - UEFI 2.3.1 Compliance - Increased VF count in SR-IOV (was 16 VFs per port, is now 32 VFs per port) - Support for next generation 10GbaseT PHY - Expose IBM Feature On Demand licensing information in UEFI configuration utility - IBM Universal Configuration Manager for Ethernet - IBM Unified Fabric Protocol - Ability to assign Switch Agnostic VLAN IDs without intermediate reboot - Changes to Personality or Multichannel in PXESelect allows further changes to be made w/o reboot - Support for IBM Flex System Fabric CN4093 10Gb Converged Scalable Switch
Known Issues: - HW iSCSI offload is not currently supported in UFP mode
Firmware Version: 4.1.422.8 Supported On: IBM Flex
Problems Fixed: - PHY Loopback fails on 10G ports - RH 6.2 and W2k8 servers crashed after shut/no shut the switch port with FCoE connection and 50 vlans configured - DSA: NIC physical loopback test fails in Command and GUI mode - Firmware update fails to update all VPD Code Level fields - DCBX does not converge with FW 4.1.422 on Cicso 5k.
Interoperability Since Previous Release: - Support for IBM Flex System x440 10Gb Virtual Fabric LOM
Firmware Version: 4.1.422.0 Supported On: IBM Flex
Problems Fixed: - SR IOV: Unable to ping IPV6 from DUT VF to Peer VF - Reading VPD from BE3 timesout making SR IOV fail in newer kernels - 1Gb devices are being suppressed by kernel during the PCI probe in SLES11 SP2 - MAC Loopback fails when Mgmt Adapter Creation fails - Failure of MAC loopback test after firmware update and warm reboot - Suppress efat dump mailbox commands in the efat dump log - Running ethtool with BE3 takes longer than expected after cable pull - Out of Order CQEs - TRACE_LOG function not printing the messages for some of the VF's - esx4/5 issue. svc target cannot be configured properly - RSCN is not generated by switch when a driver restarts - Port Link stays up even after switch removed
Interoperability Since Previous Release: - Support for IBM Flex System EN4054 4-port 10Gb Ethernet Adapter - Support for IBM Flex System CN4054 10Gb Virtual Fabric Adapter - Support for IBM Flex System x240 10Gb Virtual Fabric LOM - Support for IBM Flex System Manager 10Gb Virtual Fabric LOM
Firmware Version: 4.2.433.3 Supported On: IBM System x and BladeCenter
Problems Fixed: - Intermittent link status with IBM Virtual Fabric Switch when set to IBM Virtual Fabric Mode - The option to enable/disable SRIOV on HS23 LOM Port 0 is missing from OneCommand Manager. This selection is available in UEFI F1 setup. - Under high network stress, NMI errors can be seen with x3690 X5 and x3850 X5 - 10GbE ports running at 1GbE speed 10% below wirespeed at certain block sizes. - Unable to perform PXE boot with LPVID assigned to vNIC interface - 1Gb ports of HS23 do not handle tagging properly if multichannel mode enabled (pnic ok). Note: ESX driver patch also required. - Port flapping link after FW upgrade and "reboot" when connected to 10Gb PTM module (DAC)
Interoperability Since Previous Release: - Support for Emulex Virtual Fabric Adapter Advanced (BE2) for IBM BladeCenter - Support for Emulex Virtual Fabric Adapter (BE2) for IBM BladeCenter - Support for Emulex Virtual Fabric Adapter (BE2) for IBM System X
Known Issues: - All 10Gb KX4 ports must be used with 10Gb switches. They will not link at 1Gb. This applies to HS23 LOM and IBM BladeCenter VFA adapters only. - IPv6 IP address not visible via IMM GUI for IBM Option 90Y6456
Firmware Version: 4.2.412.0 Supported On: IBM System x and BladeCenter
Problems Fixed: - All VFS on Port 1 on optional Card unreachable from outside the host (x3750 M4) - 'Logical Admin Link' not enabled after applying BOFM in Switch Agnostic mode - IBM pre-boot Diagnostics NIC physical loopback test fails in Command and GUI mode - Multiple cosmetic issues with iSCSI UEFI BIOS - 10Gb ports do not detect 'Link Down' with some switches - 1Gb ports on HS23 LOM not able link up with IBM ICPM (FRU part 44W4486) in BCS chassis - Performance issues with specific DAC cables and SND top of rack switch
Interoperability Since Previous Release: - UEFI 2.3.1 Compliance - Increased VF count in SR-IOV (was 16 VFs per port, is now 32 VFs per port) - Support for next generation 10GbaseT PHY - Expose IBM Feature On Demand licensing information in UEFI configuration utility
Known Issues: - All 10Gb ports must be used with 10Gb switches. They will not link at 1Gb. - IPv6 IP address not visible via IMM GUI for IBM Option 90Y6456 - Intermittent link status with IBM Virtual Fabric Switch when set to IBM Virtual Fabric Mode - The option to enable/disable SRIOV on HS23 LOM Port 0 is missing from OneCommand Manager. This selection is available in UEFI F1 setup.
Firmware Version: 4.1.455.6 Supported On: IBM System x and BladeCenter
Problems Fixed: - Fixed an issue where ASIC randomly hangs losing network connectivity when certain commands are sent from the host. Resulting behavior varies between OS's (Yellow bangs in Windows, PSOD in VMWare) - PHY Loopback fails on 10G ports - UE and hang on LOM or Robalo when powering off Janice in UEFI or PXESelect - Loopback on HS23 now functional in DSAp - DCBX does not converge with firmware on a FCoE fabric - Control C Broken on BE2 - 7 second delay in GET_CARD_ATTRIBUTES command - Emulex VF adapter FCoE link lost on VMware 5.0 - flow control set for 10G Base T phys for autonegotiation - Dynamic link update is not working with NIC UEFI driver - Remove unused definitions in UEFI NIC - Emulex driver expects that the Login Options field be zero - "map r" displays a iSCSI BLK device when no LUNs presented - "X/x Exit without save" does not work as expected - Display proper description as part of EFI Driver Model Related Information - Intermittent "Configuration changed. reset to apply it now?" - Garbled output being displayed at the LUN configuration section. - All Emulex Configuration settings get lost after a reset - Perform additional checks when no Luns are mapped to the target - CHAP name only accepts 6 to 10 characters - Hii ???iSCSI Initiator Configuration??? showing old version screen - iSCSI name is case sensitive - iSCSI HII allows user to select first 8 LUNs as bootable. - iSCSI Boot Enable/Disable option should be port specific - LUN Configuration shows LUN 0, even if no LUNs presented to initiator.
Interoperability Since Previous Release: - Support for Emulex Virtual Fabric Adapter Advanced (CFFh, 2+2) for IBM BladeCenter
Firmware Version: 4.1.334.47 Supported On: IBM System x and BladeCenter
Problems Fixed: - NCSI transmit workaround for IBM - Issues with data reporting when BOFM enabled - Improved network performance throughput - Issues with IBM Management network performance - Logical link status for IBM mezz card shown as 'N/A' - Serial Over LAN fails when more than one blade installed with a single switch - Removed UEFI menu items for unsupported functions on a port - Issues with diagnositics in virtual NIC/UMC mode - Issues with MAC and PHY loopback on OneConnect devices - Several issues with FCoE on some OEM fabric switches
Interoperability Since Previous Release: - Support for HS23 Virtual Fabric LAN on Motherboard - Support for VFA III (95Y3762, 95Y3768), 90Y6456 - Support for IBM's Features on Demand (FOD) - Formal support for SR-IOV
Firmware Version: 4.0.1062.0 Supported On: IBM System x and BladeCenter
Problems Fixed: - iLab Tucson - lpfc Device Resets and path failures continue after error inject. - Writes do not complete when running LIP test on BE3 port 2 - IPV6 implementations no longer experience an unrecognized routing type error. - VLAN tagging is now disabled on the OneConnect UCNA when multichannel is disabled. - UEFI/BIOS-assigned LPVIDs now function properly on the OneConnect 10102-based UCNA. - OneConnect UCNAs no longer send and receive only double-tagged packets and discard all other packets. - PHY, MAC, and External loopback tests no longer fail on all platforms. - Firmware now returns the proper OneConnect UCNA model number. - IPV6 implementations no longer experience an unrecognized routing type error. - Installing Windows 2008 or Windows 2008 R2 on remote iSCSI disks no longer causes the system to hang. - Issues no longer occur when resetting a server configured for multipathing on VMware ESX systems. - VLAN tagging is now disabled on the OneConnect UCNA when multichannel is disabled. - UEFI/BIOS-assigned LPVIDs now function properly on the OneConnect 10102-based UCNA. - OneConnect UCNAs no longer send and receive only double-tagged packets and discard all other packets. - An adapter hardware error no longer occurs when running I/O with LUNs on x86 platforms. - Changing from Convergence Enhanced Ethernet (CEE) to Chassis Internal Network (CIN) and then rebooting no longer fails to link-flap (CIN Fabric Login (FLOGI) now occurs). - PHY, MAC, and External loopback tests no longer fail on all platforms. - Firmware now returns the proper OneConnect UCNA model number. - DCB settings can now be configured. - (iSCSISelect) When enabling DHCP for an iSCSI initiator, the Emulex iSCSI Network Configuration utility now waits approximately 20 seconds to make a DHCP connection before returning control. - (FCoE BootBIOS) FCF CEE parameters for Boot from SAN configuration are now saved properly.
Interoperability Since Previous Release: - Added support for switch-independent multichannel mode. - Added support for IBM Virtual Fabric 10Gb RackSwitch firmware 6.8.1.0. - Added support for IBM Virtual Fabric 10G Switch Module firmware 6.8.1.0. - Added support for Fibre Channel over Ethernet (FCoE) on rack systems. - Added support for iSCSI hardware offload on rack systems. - Added support for Cisco Nexus 5000-series switches.
Adapters Supported: ServeRAID M5115 SAS/SATA Controller for IBM Flex System (90Y4390)
NOTE TO SERVICE - Reference RETAIN #N/A
Version 68-1 - Suggested ------------ Firmware version 68 - Fixed issue that inadvertently shuts system off when the PSoC update is initiated. - Added I2C pass-through isolation logic during standby power - Fixed issue that inadvertently surfaces a Mezz fault message after performing a Virtual Reseat. - Added "M5115" signature to I2C 0xB0 register - Added support for RAID Link.
Version 63 - Initial Release ---------- ibm_fw_psoc3_m5115-68-1_windows_32-64_chg
Matrox G200eR (Renesas) Display Only Driver for Windows Server 2012
Rev. 4.00.01.004 - Certified with WHCK Build (8400)
KNOWN ISSUES ------------ - None
HISTORY ------- Rev. 4.00.01.003 - Beta version mtrox_dd_video_4_00_01_004_winsrvr_64_chg
ibm_dd_sraidmr_6_702_07_00-1_windows_32-64_chg
Broadcom Firmware Update Utility for IBM servers and options using the Broadcom NetXtreme and NetXtreme II ethernet adapter.
Version 2.4.1d4, Sept 2nd, 2014
This package supports the following NetXtreme I devices: ======================================================== 49Y7900 (asic => 5718)
This package will update NetXtreme I device firmware to the following levels: ============================================================================= Boot code 1.60, NCSI code 1.2.42, IMFW code 1.08, fcode 7.0.1 (0170), iSCSI code 7.10.3, *EFI code .96, PXE code 16.4.3
This package supports the following NetXtreme II devices: ======================================================== 88Y5921 (BCM57840), 94Y5230 (BCM57810), 94Y5232 (BCM57810), 94Y5181(BCM57810), 44T1361(BCM57810), 00D9700(BCM57811)
This package will update NetXtreme I device firmware to the following levels: ============================================================================== MFW 7.10.20 , L2 code 7.8.05, iSCSI code 7.10.4, EFI 7.10.21, PXE code 7.10.5