1. Change History 2. Known Issues 3. Notices 4. Contacting Support
1. Change History
The following changes have been made to the Converged Network Adapter STOR miniport driver for version 9.1.12.10.
* API version 1.28.0.117 (QLSDM) * [ER134003]: Added mailbox timeout error reset delay Resolution: Add mailbox timeout error delay count of three before reset Scope : 8200 Series Adapters * [ER132526]: Target port not login after FLOGI following path removal Resolution: Clear RSCN information on processing SNS timeout error Scope : All FCOE Adapters * [ER133640]: Flash write issue over the FCoE interface Resolution: Fixed flash write enable, write complete wait issue Scope : 8200 Adapters * [ER116106]: Handling CRC and Recovered Error on same exchange Resolution: Added support for handling CRC error status during recovered error check condition Scope : All FCOE Adapters * [ER132635]: I/O quiescence on tape device reboot Resolution: Added ratov related delay on issuing relogin after logout port update AEN Scope : All FCOE Adapters
Version 9.1.11.14
* API version 1.28.0.112 (QLSDM) * [ER114980]: Need to determine which active image to execute (primary or secondary) Scope : 8400 Series Adapter Resolution: Added support in driver * [ER115248]: Read NVRAM fails after write Resolution: Added delay between write and read in api Scope : 8200 Series Adapters * [ER115805]: Failed WHQL Resolution: Not to load fw if already loaded Scope : 8400 Series Adapters * [ER114713]: Poor read write performance at certain large block sizes Resolution: Fixed delay in handling SRB queue Scope : 8300 Series Adapters * [ER116203]: FCoE capturing firmware minidump despite the presence of iSCSI driver Resolution: Added check for active iSCSI func before reset recovery Scope : 8400 Series Adapters * [ER116206]: FCoE driver does not capture minidump after three peg halts Resolution: Reset firmware dump limit counter on firmware dump retrieval Scope : All FCOE Adapters * [ER116177]: Win2k12 R2 crashes during SNAP6 online update with baseline added. Resolution: Fixed certain IOCTL pass-through command sense data pointer on W12 extended Srb Scope : 8400 and 8200 Series Adapters * [ER131902]: FDMI-2 support Resolution: Added FDMI-2 support Scope : 8300 Series Adapters * [ER132037]: Flash fw version not updated Resolution: Add delay between flash write and flash read in api Scope : 8200 Series Adapters * [ER132112]: Echo ELS passthru support Resolution: Added Echo ELS passthru support Scope : 8200 and 8300 Series Adapters
Version 9.1.11.13
* [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 * [ER111553]: Failed to push driver version in OCBB space Scope : 8200 Series Adapter Resolution: Skipped issuing get param mbx * [ER114741]: Login to tape device times out Scope : All FC and FCOE Adapter Resolution: Retry Relogin once * [ER113924]: Feature request to add get/set SerDes registers Scope : 8400/2700 Series Adapter Resolution: Added api interface * [ER112258]: Peg4 abort (iocbs not free) while running driver load-unload with 64 VPs per port Scope : 8400 Series adapter Resolution: Added check for HBA stop state before response queue processing
Version 9.1.11.12
* API version 1.28.0.111 (QLSDM) * [ER110756]: Incorrect sense data debug message during certain check condition status response Scope : All FCOE Adapters Resolution: Fixed debug message output with data based on sense error code * [ER108790]: FCoE Internal Loopback does not complete Scope : 8300 and 8400 Series adapter Resolution: Wait for IDC to complete loopback ioctl * [ER111292]: Incorrect CDB pointer type used on Windows Server 2012 extended SRB Resolution: Fixed CDB pointer type declaration for extended SRB * [ER111432]: Unable to create Vports when Microsoft vFC's are existing Scope : All FCOE Adapters Resolution: Check the active VP count in the first half of VP database * [ER111645]: Update nvram fails read-back verify Scope : 8200 Series adapter Resolution: Added delay before read in QLSDM
Version 9.1.11.11
* API version 1.28.0.110 (QLSDM) * HotFwLoad is no longer supported because it causes whql chaos failure witb mb error [ER109340] * Added support for user configurability of maximum number of NPIVs per adapter port [ER 109407] * Fixed issue with device init timeout value on 8200 Series adapter [ER108082] * Added device state check before hardware access for 8200 Series adapter [ER107969] * Added 8200 Series adapter IO timeout handling for pass-through command [ER107784] * Fixed number of IOCB entries required for large transfer [ER107785] * Added more registry changes to set number of vps and number of targets per vp [ER107100] * Enabled force fw dump support in api [ER107301] * Fixed io completion and blocking issue during quiscent state for 8200 Series adapter [ER107135] * Fixed io completion issue on reset failure for 8200 Series adapter [ER107148] * Fixed rom lock issue during firmware minidump for 8200 Series adapter [ER107368] * Removed invalid mailbox command error detection for 8200 Series adapter [ER107368] * Change OCBB implementation to use mailbox register method for 8200 Series adapter [ER106341] * Added delay before writing flash sector to address flash corruption [ER106427] * Added delay before reset to address firmware hang [ER106427] * Added registry strings to set number of vps and number of targets per vp [ER107100] * Fixed empty firmware image version due to 8200 Series adapter (based on latest controller revision) not regconized by driver [ER105290] * Fixed unpack macro warning in QLSDM [ER105492] * Fixed VP creation failure because of dup wwpn in the switch [bug 25373] * Added support in qlsdm to support new name of inbox driver [ER105740] * Added support for Long Range [ER0000000110404]
Version 9.1.11.10
* API version 1.28.0.106 (QLSDM) * Fixed beaconing for 8300 Series adapter according to the new firmware spec [ER101442] * Added api to read Hilda ASIC temperature [ER102430] * Added mailbox passthru api [ER102988] * Added reading ASIC temperature for Ninja and Sensei [ER103001] * Allow application modify "Execution Throttle" field in nvram for 8300 series adapters [ER103297] * Add Service Pack number to OSName [ER104204] * Removed Advanced tab from Device Manager properties * Fix MailboxPassthru by removing restriction on writing mbx8 register [ER104361]
Version 9.1.10.17
* API version 1.28.0.101 (QLSDM) * Added 8300 series Adapter temperature handling [ER94159] * Added support for 8300 series Adapter (IDC, minidumps (as applicable), loopback, MCTP, mngt api) [ER92242] * Driver needs to determine, depending on the config, if it needs to send echo or external (8300 series Adapter FCoE) loopback command[ER94711] * Added 8300 series Adapter FCoE IDC driver presence bit support [ER95252] * Added 8300 series Adapter FCoE IDC driver reset/quiescent ack support [ER95250] * Returned NOT_SUPPORTED for SDForceFcMctpDump in api [ER96352] * Removed unnecessary mailbox retry on 8300 series Adapter NIC init state [ER96652] * Added adapter disable check on timer routine re-init state handling [ER96662] * Fixed NIC init state issue on 8300 series Adapter NIC Firmware State Change AEN [ER97159] * Fixed BSOD running FCoE Internal loopback when mailbox set config fails [ER97406] * Changed 8300 series Adapter FCoE NIC init state handling on 0x8200 AEN [ER97159 ] * Added registry to control IO timeout event log/sense data report for Win12 [ER98501] * Ported MPS patch from 8100 to 8300 series Adapter to fix BFS win12 installation BSOD [ER96750] * Removed event logging for fw dump event [ER98897] * Limit MSI-X resources to 2 for P3P [ER97967] * Use macros to access SRB and SRB_EXT for W12 [ER097387][ER98974] * Fixed crash in fwdump routine when 8100 series Adapter dumps fwdump [ER101039] * Fixed api to reload firmware on second port after firmware is flashed on first port for 8100 series Adapter [ER101118]
Version 9.1.10.16
* API version 1.28.0.97 (QLSDM) * Fixed extra 4K in variable transfer sizes (64K, 128K) [ER94081] * Fixed memory access error [part of ER96188] * Added support to relogin if the device logs out [ER93867]
Version 9.1.10.15
* Added Windows Server 2012 support [ER92249] * Fixed Flash Layout Table memory usage [ER93668] * API version 1.28.0.96 (QLSDM) * Add SDGetHbaDeviceNodePropertyResync in QLSDM [ER92842] * Add cached FLT support [ER93468] * FCoE driver doesn't refresh FW versions when flash is updated through NIC on 82xx [ER85860] * 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] * P3+ Minidump buffer pointer issue [ER95253] * FCoE driver doesn't refresh FW versions when flash is updated through NIC on 82xx [ER91115]
Version 9.1.9.39
* Let NIC driver perform reset if NIC driver is present * FCoE driver stuck in DPC for too long causing assertion during peghalt testing [ER89373] * Updating Qlogic Firmware gives Read only Filesystem Error [ER86864]
Version 9.1.9.38
* Addressed issue related to peg halt tests
Version 9.1.9.37
* System unresponsive after FW hang [ER85860] * API version 1.28.0.90 (QLSDM)
Version 9.1.9.36
* Log halt events and event 800267xy for ipe * Default to disable minidump * Improved sem5 polling for reset * QCC CLI hung after making personality change on QLE8242 [ER83352] * Cleaned up accumulated io completion properly when link down occurs * CNA 81xx can not load the driver after hot plug in OEM system [ER83352] * Assertion failures and system hang when peg halt is done in loop with NIC and FCoE driver running. [ER83352] * API version 1.28.0.90 (QLSDM)
Version 9.1.9.19
* Added support and retry logic code to track down io failure during link down.
Version 9.1.9.18
* Made compatible with OEM DSM that handles only path failure by deferring io error when link/port down timeout exceeds.
Version 9.1.9.17
* Added new cold reset when loaded before nic driver during IBFT. * Synchronized vp's rscn handling with vp 0.
Version 9.1.9.16
* Target discovery not completed within port down timeout would cause driver to logout the first discovered target. [ER78315] * Target discovery was initiated prior to firmware ready. [ER78316] * The firmware version in QCC is not updated correctly after flashing using NIC interface. [ER78827] * API version 1.28.0.83 (QLSDM) * Firmware v5.04.04
Version 9.1.9.15
* Added support for the 82xx family of Converged Network Adapters * API version 1.28.0.80 (QLSDM) * Firmware v5.04.01
Version 9.1.8.27
* Degraded write performance with large file and Windows Server 2008 R2 file system [ER71241] * API version 1.28.0.73 (QLSDM) * Firmware v5.03.02
Version 9.1.8.26
* Added LR Converged Network Adapter Adapter support (QLE81xx) * Updated embedded firmware to v5.3.2 (QLE8042) * Addressed virtual link down issue reported by HDS (ER68836) * API version 1.28.0.73 (QLSDM) * Firmware v5.03.02
Version 9.1.8.19
* System freezes after active BNT CEE switch is rebooted * Adapter management utility fails to delete Virtual ports * API version 1.28.0.73 (QLSDM) * Firmware v5.02.01
Version 9.1.8.17
* API version 1.28.0.73 (QLSDM) * Firmware v5.01.03
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 assistance, 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 2015. 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-lnvgy_dd_fc_cna-9_1_12_10_windows_32-64_chg
Lenovo IMM2 PBI device NULL driver
Version 1.0.1.2015 ------------ - Initial release lnvgy_dd_immpbi_1_0_1_2015_windows_32-64_chg
=============================================================================== Emulex 16GB FC Firmware Package ===============================================================================
Firmware Version: 10.6.228.14 Supported On: Lenovo System x Rack and Flex
Problems Fixed: - None
Incremental Interoperability: - This software/firmware is only supported with Lenovo Branded System BIOS - xClarity firmware download on specific platforms (requires minimum IMM version-please consult Lenovo on the co-requisite)
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: 10.4.303.3 Supported On: System x Rack
Problems Fixed: - VF speed is showing wrongly in ethtool and Ocm. - Firmware download using ethtool -f command is failing with opcode 172-1 failed:status 100-208 - Can not set the DelayDeviceDiscovery to Max (255) by ASU tool - Ports intermittently logging in to Brocade switch as G-Port
Incremental Interoperability: - xHMC event notifications on specific platforms (requires minimum IMM version-please consult Lenovo on the co-requisite) - Lenovo branded versions of formerly IBM branded products
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)
Problems Fixed: - Blue screen after firmware flash to with virtualized WWPN
Incremental Interoperability: - xHMC event notifications on specific platforms (requires minimum IMM version-please consult Lenovo on the co-requisite) - Lenovo branded versions of formerly IBM branded products
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)
Incremental Interoperability: - Grantley based ITE
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: 10.2.377.18 Supported On: System x
Problems Fixed: - Link does not recover during IO + Port Cycling with FW 1.1.43.3 - BIOS versions return multiple values - LPe16k point to point (direct attached) connection not bootable in legacy BIOS - LPe16k point to point (direct attached) connection performance issues in legacy BIOS
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: 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 - Reported firwmare version in CMM is intermittently incorrect - Unable to set pause Flow control when connected to DCBX switch with PFC off - during cable pull testing a 40 second delay in link up experienced with Brocade switch ports - I/O completion and abort completion WQE are generated out of order - Unexpected WCQE in RQ CQ seen in RXRI mode while performing cable pulls
Incremental Interoperability: - Inventory support for the Flex xHMC (Chassis Hardware Management Console) - Crosslink & Express Lane on LPe16k based solutions
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
Integrated Management Module II (IMM2) Firmware Update Change History
Version 1.95, 10/30/15, TCOO12R-NonCritical - Supports x240 M5 , Machine Type 9532, 2591 - Supports x240 M4, Machine Types: 7162, 2588 - Supports x440, Machine Type 7167, 2590 - Supports x880 X6, Machine Type 7196,4258 - Supports x3750 M4, Machine Type: 8753 - Supports x3850 X6, Machine Type: 6241 - Supports x3550 M5, Machine type 5463 - Supports x3650 M5, Machine type 5462 - Supports x3500 M5, Machine type 5464 - Added a description in the CLI interface explaining what the threshold headers represent. - Enhanced the CLI vpd command to include firmware information for adapters - Enhanced the CLI 'syshealth' command to include the hardware health - Extended the IPMI disable/enable feature to all interfaces - PSU over-subscription support with NM3.0 hard capping on x3500 M5/x3550 M5/x3650 M5 - Added support to report both rated and actual DIMM speeds - Added CPU SKU and model information to the data that IMM2 shows for each CPU - Added inventory support for Lenovo XClarity Administrator - Added command to show power statistics in the Command Line interface - Added support for updating the administrator password by ASU - Added support to configure the remote control¡¥No response time interval¡¦setting using ASU - Firmware changes to address security vulnerabilities: CVE-2013-7423, CVE-2015-0286, CVE-2015-0287, CVE-2015-0289, CVE-2015-0292 CVE-2015-0293, CVE-2015-0209, CVE-2015-0288, CVE-2015-4000, CVE-2015-1789 CVE-2015-1790, CVE-2015-1792, CVE-2015-1791, CVE-2014-8176, CVE-2015-1799 CVE-2014-7817, CVE-2014-9402, CVE-2015-1472, CVE-2015-1788, CVE-2015-5600 CVE-2013-2207, CVE-2014-8121, CVE-2015-1781, CVE-2015-5722 - Fixed a problem on x3500 M5 where the IMM2 reports the wrong HDD drive slot number - Fixed a problem where the IMM2 Web content did not display fully - Prevent anonymous users from uploading files - Corrected the IMM2 MIB to include the IPMI configuration node to the tree - Corrected an LED Name mismatch between Lenovo XClarity Administrator and the IMM2 Web interface - Fixed a problem on the Flex Compute Node where the IMM2 login failed when the CMM is in NIST 800 mode - Fixed a problem where that the CapsLock does not work when using the IMM2 remote console with Japanese OS
Version 1.81, 7/14/15, TCOO10Q - NonCritical - Supports x3650 M5, Machine type 5462 - Added io3 Flash Adapter thermal management enhancements to avoid system shut down. - FPGA update to fix a problem that prevents the system from booting when no ML2 card is installed - Corrected an issue where the firmware falsely issues a critical alert when riser cards are installed without a fan.
Version 1.72, 5/29/15, TCOO10H - Critical - Supports x3650 M5, Machine type 5462 - Supports x3550 M5, Machine type 5463 - NOTE: To avoid possible exploits of the vulnerabilities discovered with SSLv3 encryption, support for SSLv3 has been removed from the IMM2 firmware. To access the IMM2 web interface using https, users must make use of client browser versions that support TLS1.2 encryption. Refer to RETAIN Tip H214508 for details. - Added support for the June 2015 leap second adjustment.
Version 1.72, 5/29/15, TCOO10H - Critical - Supports NeXtScale nx360 M5, Machine type 5465 - NOTE: To avoid possible exploits of the vulnerabilities discovered with SSLv3 encryption, support for SSLv3 has been removed from the IMM2 firmware. To access the IMM2 web interface using https, users must make use of client browser versions that support TLS1.2 encryption. Refer to RETAIN Tip H214508 for details. - Firmware changes to address security vulnerabilities: CVE-2015-2808, CVE-2015-0204, CVE-2014-3566 - Fixed IMM responsiveness while SD RAID Media Adapter is installed in the system - To support new security requirements the IMM firmware now requires the Java remote presence client to be at version 1.8 or newer - Fixed a PXE Boot issue that occurred when the PXE boot is requested on the IMM web page - Added copy/paste support for certificate generation. - Added support for the time zone changes in Russia - Updated the IMM web server to support Internet Explorer 8-11, Firefox 3.6-29, and Chrome 13-35 - Added web interface support for the SD RAID Media Adapter - Corrected the firmware that erroneously indicated an ME error - Added support for the June 2015 leap second adjustment.
Version 1.72, 5/29/15, TCOO10H - Critical - Supports NeXtScale nx360 M5 DWS, Machine type 5467 - NOTE: To avoid possible exploits of the vulnerabilities discovered with SSLv3 encryption, support for SSLv3 has been removed from the IMM2 firmware. To access the IMM2 web interface using https, users must make use of client browser versions that support TLS1.2 encryption. Refer to RETAIN Tip H214508 for details. - Added support for the June 2015 leap second adjustment.
Version 1.71, 5/05/15, TCOO10D - Critical - Supports x3650 M5, Machine type 5462 - Supports x3550 M5, Machine type 5463 - Firmware changes to address security vulnerabilities: CVE-2015-2808, CVE-2015-0204, CVE-2014-3566 - Fixed IMM responsiveness when SD RAID Media Adapter for System x is installed and un-configured - To support new security requirements the IMM firmware now requires the Java remote presence client to be at version 1.8 or newer - Fixed a PXE Boot issue that occurred when the PXE boot is requested on the IMM web page - Added copy/paste support for certificate generation. - Added support for the time zone changes in Russia - Updated the IMM web server to support Internet Explorer 8-11, Firefox 3.6-29, and Chrome 13-35 - Added web interface support for the SD RAID Media Adapter - Added PCIe device link down detection - Fixed a problem that System will be shut down due to abnormal PTU result
Version 1.70, 4/30/15, TCOO10C - NonCritical - Supports NeXtScale nx360 M5 DWS, Machine type 5467 - Firmware changes to address security vulnerabilities: CVE-2015-2808, CVE-2015-0204, CVE-2014-3566 - Fixed IMM responsiveness while SD RAID Media Adapter is installed in the system - To support new security requirements the IMM firmware now requires the Java remote presence client to be at version 1.8 or newer - Fixed a PXE Boot issue that occurred when the PXE boot is requested on the IMM web page - Added copy/paste support for certificate generation. - Added support for the time zone changes in Russia - Updated the IMM web server to support Internet Explorer 8-11, Firefox 3.6-29, and Chrome 13-35 - Added web interface support for the SD RAID Media Adapter - Corrected the firmware that erroneously indicated an ME error
Version 1.64, 4/27/15, TCOO08V - NonCritical - Supports the System: x240 , Machine Types: 7162, 2588 - Supports the System: x240 M5 , Machine Type 9532, 2591 - Supports the System: x440, Machine Type 7167, 2590 - Firmware changes to address security vulnerabilities: CVE-2015-2808, CVE-2015-0204, CVE-2014-3566, CVE-2014-3511, CVE-2014-3509, CVE-2014-3505, CVE-2014-3506, CVE-2014-3508, CVE-2014-5139, CVE-2014-3507, CVE-2014-3510, CVE-2013-4332, CVE-2014-3513, CVE-2014-3567, CVE-2014-3568, CVE-2014-3569, CVE-2014-3570, CVE-2014-3571, CVE-2014-3572, CVE-2014-8275, CVE-2015-0205, CVE-2015-0206, - Fixed a cooling issue when the Fusion IO adapter is present - Fixed IMM responsiveness when SD Media Adapter for System x is installed and un-configured - To support new security requirements the IMM firmware now requires the Java remote presence client to be at version 1.8 or newer - Fixed an issue where the inventory on a Flex compute node is not always updated in the CMM - Fixed a PXE Boot issue that occurred when the PXE boot is requested on the IMM web page - Added support for the ServeRAID M5215 with 2GB Flash Enablement - Flex System x240 M5 compute Node - Added copy/paste support for certificate generation. - Added support for the time zone changes in Russia - Updated the IMM web server to support Internet Explorer 8-11, Firefox 3.6-29, and Chrome 13-35
Version 1.60, 4/16/15, TCOO08Q - NonCritical - Supports the System: x240 , Machine Types: 7162, 2588 - Firmware changes to address security vulnerabilities: CVE-2015-2808, CVE-2015-0204, CVE-2014-3566, CVE-2014-3511, CVE-2014-3509, CVE-2014-3505, CVE-2014-3506, CVE-2014-3508, CVE-2014-5139, CVE-2014-3507, CVE-2014-3510, CVE-2013-4332, CVE-2014-3513, CVE-2014-3567, CVE-2014-3568, CVE-2014-3569, CVE-2014-3570, CVE-2014-3571, CVE-2014-3572, CVE-2014-8275, CVE-2015-0205, CVE-2015-0206, - Fixed a cooling issue when the Fusion IO adapter is present - Fixed IMM responsiveness when SD Media Adapter for System x is installed and un-configured - To support new security requirements the IMM firmware now requires the Java remote presence client to be at version 1.8 or newer - Fixed an issue where the inventory on a Flex compute node is not always updated in the CMM - Fixed a PXE Boot issue that occurred when the PXE boot is requested on the IMM web page - Added support for additional models of the Flex System x880 Compute Node and the Flex System x240 Compute Node - Added support for the ServeRAID M5215 with 2GB Flash Enablement - Flex System x240 M5 compute Node - Added copy/paste support for certificate generation. - Added support for the time zone changes in Russia - Updated the IMM web server to support Internet Explorer 8-11, Firefox 3.6-29, and Chrome 13-35
Version 1.51, 3/23/15, TCOO08K - Suggested - Supports NeXtScale nx360 M5, Machine type 5465 - Supports NeXtScale nx360 M5 DWS, Machine type 5467 - Changed the firmware to handle erroneous temperature readings from the nVidia Tesla K80 GPU card instead of shutting the system down
Version 1.50, 3/2/15, TCOO08I - Suggested - Supports NeXtScale nx360 M5, Machine type 5465 - Supports NeXtScale nx360 M5 DWS, Machine type 5467 - Supports SD Media Adapter for System x - Fixed reporting of incorrect Active Energy Management values
Version 1.41, 2/12/15, TCOO08J - Critical - Supports System x3500 M5, Machine type 5464 - Added thermal management enhancements for the System x3500 M5 - Fixed a problem on the System x3500 M5 where on rare occasions after an AC cycle, the IMM is unable to start up and the heartbeat LED remains off
Version 1.40, 1/19/15, TCOO08E - Initial Release - Supports System x3500 M5, Machine Type: 5464
Version 1.20, 12/12/14, TCOO04I - Initial Release - Supports Flex System x440 Compute Node, Machine Types: 7167 and 2590 - Supports Flex System x240 M5 Compute Node, Machine Types: 9532 and 2591
Version 1.08, 2/03/15, TCOO06A - Initial release - Supports System x3850/x3950 X6, Machine Type: 6241
Version 1.06, 2/03/15, TCOO06B - Initial Release - Supports System x3750 M4, Machine Type: 8753
Version 1.02, 12/26/14, TCOO02J - Critical - Supports System x3500 M5, Machine type 5463 - Supports System x3650 M5, Machine type 5462 - Fixed a problem on x3550 M5 & x3650 M5 where the IMM may erroneously report system board 5V value
Version 1.01, 12/04/14, TCOO02I - Suggested - Supports NeXtScale nx360 M5, Machine type 5465 - Enabled the call home (call support) feature - Fixed a CPU throttle issue that may occur when nodes are unplugged or plugged in
Version 1.00, 12/03/14, TCOO02G - Initial Release - Supports System x3650 M5, Machine Types: 5462 - Supports System x3550 M5, Machine Types: 5463 - Supports System NeXtScale nx360 M5, Machine Type: 5465, 5467
(C) Copyright Lenovo 2015.
LIMITED AND RESTRICTED RIGHTS NOTICE: If data or software is delivered pursuant a General Services Administration "GSA" contract, use, reproduction, or disclosure is subject to restrictions set forth in Contract No. GS-35F-05925.
(C) Copyright International Business Machines Corporation 2008 - 2014. All rights reserved.
Lenovo and the Lenovo logo are trademarks of Lenovo in the United States, other countries, or both. lnvgy_fw_imm2_tcoo12r-1_95_anyos_noarch_chg
=============================================================================== Emulex NIC Device Driver for Windows ===============================================================================
Device Driver Version: 10.6.236.0 Supported On: Lenovo System x Rack and Flex
Problems Fixed: -
Incremental Interoperability: - RoCE v2 with XE-104 P2 based - Flex System CN4052 R 2-port 10Gb Virtual Fabric Adapter - Lenovo 00AG572 VFA5 2x10GbE SFP+ PCIe Adapter - Lenovo 00AG582 VFA5 2x10GbE SFP+ PCIe FCoE/iSCSI Adapter - Lenovo 00AG562 VFA5 ML2 2x10GbE SFP+ Adapter
Known Issues: - RoCE support is limited to named applications. Contact your local OEM Sales Support for more information.
- There is no benefit from using NVGRE on more than one vNIC attached to the same physical port. As such, only one vSwitch should be attached to a physical port. Attaching more than one vSwitch to a physical port could cause a dramatic decrease in performance. This is a permanent limitation
Device Driver Version: 10.4.255.23 Supported On: System x Rack
Problems Fixed: - OCe14000 with LRO enabled causes OS to reboot - Observing "IOCTL to firmware failed" in the Diagnostics tab after installing the Driver to the Virtual Function
Incremental Interoperability: - Lenovo branded versions of formerly IBM branded products
Known Issues: - RoCE support is limited to named applications. Contact your local OEM Sales Support for more information.
- There is no benefit from using NVGRE on more than one vNIC attached to the same physical port. As such, only one vSwitch should be attached to a physical port. Attaching more than one vSwitch to a physical port could cause a dramatic decrease in performance. This is a permanent limitation
Device Driver Version: 10.6.236.0 Supported On: Flex
Problems Fixed: - BSOD when booting to windows 2012 R2 with HyperV with 4x CN4054 in system - Driver reports too many VMQ resources
Incremental Interoperability: - Lenovo branded versions of formerly IBM branded products
Known Issues: - RoCE support is limited to named applications. Contact your local OEM Sales Support for more information.
- There is no benefit from using NVGRE on more than one vNIC attached to the same physical port. As such, only one vSwitch should be attached to a physical port. Attaching more than one vSwitch to a physical port could cause a dramatic decrease in performance. This is a permanent limitation
Device Driver Version: 10.2.413.1 Supported On: System x, BladeCenter and Flex
Problems Fixed: - Packet loss observed during affinity changes. During the change, a 'pause' occurs, packets will get dropped, bandwidth utilization will drop because of the stall, and overall performance will be inconsistent.
- When using multi-channel in a vSwitch environment, map one PF to the hyper-visor vSwitch and then use the other PF’s/vNICs for live migration network, FCoE/iSCSI, Management network and the like in order to get the best performance.
Incremental Interoperability: - Grantley based ITE
Known Issues: - RoCE support is limited to named applications. Contact your local IBM Sales Support for more information.
- There is no benefit from using NVGRE on more than one vNIC attached to the same physical port. As such, only one vSwitch should be attached to a physical port. Attaching more than one vSwitch to a physical port could cause a dramatic decrease in performance. This is a permanent limitation
Device Driver Version: 10.2.370.18 Supported On: System x and BladeCenter
Problems Fixed: - Network connectivity is getting lost for VMQ interfaces during VMQ CPU affinity stress test
Incremental Interoperability: - None
Known Issues: - RoCE support is limited to named applications. Contact your local IBM Sales Support for more information. - Packet loss observed during affinity changes. During the change, a 'pause' occurs, packets will get dropped, bandwidth utilization will drop because of the stall, and overall performance will be inconsistent. - When using multi-channel in a vSwitch environment, map one PF to the hyper-visor vSwitch and then use the other PF’s/vNICs for live migration network, FCoE/iSCSI, Management network and the like in order to get the best performance. - There is no benefit from using NVGRE on more than one vNIC attached to the same physical port. As such, only one vSwitch should be attached to a physical port. Attaching more than one vSwitch to a physical port could cause a dramatic decrease in performance.
Device Driver Version: 10.2.348.2 Supported On: IBM System x
Problems Fixed: - BSOD after running IO's using NVGRE offload for 16VM's, also affects VMQ and SRIOV - Option to enable/disable RSS not visible in NIC advanced properties - Network connectivity is getting lost for VMQ interfaces - Performance and Fairness for 16VM's is uneven on Windows 2012 R2
Known Issues: - RoCE support is limited to named applications. Contact your local IBM Sales Support for more information - Packet loss observed during affinity changes. During the change, a 'pause' occurs, packets will get dropped, bandwidth utilization will drop because of the stall, and overall performance will be inconsistent. - When using multi-channel in a vSwitch environment, map one PF to the hyper-visor vSwitch and then use the other PF’s/vNICs for live migration network, FCoE/iSCSI, Management network and the lie in order to get the best performance. - There is no benefit from using NVGRE on more than one vNIC attached to the same physical port. As such, only one vSwitch should be attached to a physical port. Attaching more than one vSwitch to a physical port could cause a dramatic decrease in performance.
Device Driver Version: 10.2.261.11 Supported On: IBM System x, BladeCenter, and Flex
Problems Fixed: - When MAC learning is enabled, driver should not drop internally switched packets - Incorrect Port numbering in nic multichannel tab - No link down on UMC 0 bw min and max - BE3: very low throughput transmit on SR IOV Guest OS (netperf less than 100Mbps although 10Gb link) - BE3 Win2k12R2 Hyper v server crashed during the creation of virtual switch using the LACP teamed adapter - NIC Teaming (failover/failback) fails on HS23 LOM
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 - 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.
Device Driver Version: 10.0.718.26 Supported On: IBM System x
Problems Fixed: - VMQ driver option incorrectly available for 1Gb ports of BE3 - BE2 and BE3: Hitting Nic initialization failure after updating driver - Update custom property page display for BE3R - Windows NIC driver always reports initial link status as UP
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
Adapters Supported: ServeRAID M1215 SAS/SATA Controller for Lenovo System x (46C9115)
NOTE TO SERVICE - Reference RETAIN #N/A
Version 24.9.0-0026 - High Impact/High Probability of Occurrence ------------------- Fixes: - Fixed an issue in iMR mode where "Controller encountered fatal error and was reset" errors can occur under heavy IOs and background operations (SCGCQ00894643) (RETAIN Tip H001154) - Fixed an issue where driver health protocol may report unhealthy if more than 64 VDs are configured (SCGCQ00830918) - Fixed an issue where VDs beyond 64 may not be reported in legacy OpROM pOST display list (SCGCQ00826174) - Fixed an issue where "StorCLI /cX add VDs detailed" menu provides incorrect syntax for R1E (SCGCQ00501437) - Fixed an issue where CLI "show all" command fails with error "Error! App exec failed in FW PCLIError! Internal" (SCGCQ00518543) - Fixed an issue where the dedicated hot spare option in HII may be grayed out for RAID1 under the profile based configuration menu (SCGCQ00873973) - Fixed an issue where HII drive management and enclosure management may report only port 4-7 connected when ports 0-3 and 4-7 are connected to the same expander (SCGCQ00870398) - Fixed an issue where HII may hang or multiple enclosure entries will be seen in enclosure eanagement form after repeated navigation (SCGCQ00878350) - Fixed an issue where a 15-character VD name is truncated to 14 characters (SCGCQ00854606) - Fixed an issue where profile-based VD creation menu doesn't explain each profile's drive selection criteria (SCGCQ00884229) - Fixed an issue where some overlap may be seen during legacy OpROM pOST when more than one controller is present and there are boot messages for the second controller (SCGCQ00866982) - Fixed an issue where MSM may display "The requested command has invalid arguments" when 'Schedule Consistency Check' is selected (SCGCQ00832908) - Fixed an issue where a controller may not be accessible through RAIDLink until the OS is up and running (SCGCQ00847001) - Fixed an issue where drive firmware on certain SED/FDE SATA SSDs could not be updated (SCGCQ00856977) - Fixed an issue where an error "MonTask: line 1272 in file ../../raid/1078dma.c" may occur while running IO in a virtual machine (SCGCQ00822403) - Fixed an issue where a JBOD PD may not be able to be converted to Unconfigured Good in Solaris (SCGCQ00834552) - Fixed an issue where an "Instruction TLB" error may occur when a StorCLI /cX set autopdcache=r0 or similar command is executed (SCGCQ00835586) - Fixed an issue where firmware may hang in multipath scenario when cable pull/push test is performed at the same time as an array property change is issued (SCGCQ00836496) - Fixed an issue where a secure erase on a SATA SSD in HII may fail or lead to a hang in HII (SCGCQ00841317) - Fixed an issue where the controller may reset in some situations with media errors and VD undergoing manual rebuild after drives are taken offline, under stress (SCGCQ00825126) - Fixed an issue where the 3x08 chip Errata ID - 2012-RC714 may lead to PCIE 3.0 hot reset non-compliance (SCGCQ00829129) - Fixed an issue where rebuild performance on RAID1 is always slower than other RAID configurations (SCGCQ00841359) - Fixed an issue where patrol read properties window won't show up in MSM (SCGCQ00797676) - Fixed an issue where an L2 unrecoverable cache error may be seen during patrol read properties modification when VDs are made offline (SCGCQ00847162) - Fixed an issue where a VD erase may not resume after reboot when a CacheCade VD is present (SCGCQ00851810) - Fixed an issue where controller boots to OS even when pinned cache is present and boot error handling is set to stop on error (SCGCQ00860503) - Fixed an issue where firmware cannot detect some newly-inserted drives (SCGCQ00860270) - Fixed an issue where "MonTask: line 1518 in file ../../raid/2108vI2o.c" may occur in conjunction with a VMWare Guest abrupt shutdown (SCGCQ00852670) - Fixed an issue where connector name field is displayed as port 4-7 when actually both ports 0-3 and 4-7 are connected (SCGCQ00868215) - Fixed an issue where "line 243 in file ../../raid/1078int.c" error may occur during rebuild (SCGCQ00856256) - Fixed an issue where no error message shown when there is a failure to disable SSD caching of an offline VD (SCGCQ00905310) - Fixed an issue where "1 JBOD(s) found on the host adapter" may appear in legacy OpROM post even though no JBODs are present (SCGCQ00882729) - Fixed an issue where firmware may halt with "line 245 in file ../../raid/1078int.c" in log under stress with maximum VDs and background operations (SCGCQ00890133) - Fixed an issue where the HII may report a different Battery/SuperCap status than MSM/StorCLI would (SCGCQ00879726) - Fixed an issue where the firmware may hang due to an "Illegal Instruction Exception" error (SCGCQ00850441) - Fixed an issue with less-than-expected performance of Write-Verify SCSI commands in RAID0 (SCGCQ00866261)
Enhancements: - Added more descriptive port information to include link width in PHYs (SCGCQ00854621)
Version 24.7.0-0056 - High Impact/High Probability of Occurrence ------------------- Fixes: - Fixed issue where UNMAP or TRIM commands might fail on SSD in JBOD Mode (SCGCQ00853147) (RETAIN Tip H214381)
Version 24.7.0-0052 - High Impact/High Probability of Occurrence ------------------- Fixes: - Incorporated MegaRAID FW stack 6.6 - Fixed an issue where a drive reporting itself as 0MB shows up as Unconfigured Good (SCGCQ00784319) - Fixed an issue where the firmware might hang during error handling of sense code 0b/47/xx (SCGCQ00701744) - Fixed an issue where some SSDs may cause the system to hang at POST (SCGCQ00595506) - Fixed an issue where R0, R5 and R6 read bandwidth dropped abruptly at 256K block size with 24drive SATA SSD config (SCGCQ00609808) - Fixed an issue where RAID10 performance is under-performing (SCGCQ00794773) - Fixed an issue where in some cases the LSI chip would convert an incoming chip reset to a system reset (SCGCQ00664261) - Fixed an issue where many pRdm/pRequest/pReply messages would appear in LSI firmware log (SCGCQ00752195) - Fixed an issue where when a consistency check is aborted, the adapter may reset (SCGCQ00819811) - Fixed an issue where Patrol read does not start and gives incorrect warning message (SCGCQ00728404) - Fixed an issue where the controller may reset unexpectedly during error handling (SCGCQ00825126) - Fixed an issue where StorCLI may report Enclosure Not Found (SCGCQ00843146) - Fixed an issue where x880 nodes may report excessive 'SCL STUCK Low! While I2C Rx Receiving Data' in LSI firmware log (SCGCQ00728935) (Retain TIP H213891) - Fixed an issue where the command timeout for FED (SED) drives is too short, resulting in 05/2C codes (SCGCQ00797355) - Fixed an issue where bandwidth falls as queue depth increases on R5/R6 24/40 drive SAS HDD configs at 256k block size (SCGCQ00701768) - Fixed an issue where some SSDs may get bad PI CRC even though data is correct resulting in drive failure (SCGCQ00819314) - Fixed an issue where the firmware may exposed more than 240VDs to host even if host driver doesn't support that many (SCGCQ00829214) - Fixed an issue where MSM may display an error message if 'Schedule Consistency Check' is selected (SCGCQ00832908) - Blocked CacheCade deletion when multiple CacheCade volumes are associated with Virtual Drives (SCGCQ00838953) (Retain TIP H214323) - Fixed an issue where the MegaRAID POST may display "The native configuration is not supported by current firmware" (SCGCQ00812842) - Fixed an issue where the EFI Driver Health Protocol will show Fail when more than 64VDs are configured (SCGCQ00830918) - Fixed an issue where OS may BSOD when upgrading firmware on a controller the OS has booted from if Online Controller Reset is performed (SCGCQ00825725) - Fixed an issue where controller wouldn't go to halt state after multibit ECC errors detected (SCGCQ00783186) - Fixed an issue where Opal-compliant FDE (SED) drives were being marked as FDE/SED when they shouldn't be (SCGCQ00788250) - Fixed an issue where correctable ECC error interrupts were being masked (SCGCQ00814997) - Fixed an issue where reconstructed VDs were not being reflected in MSM until system reboot (SCGCQ00839018) - Fixed an issue where controller commands may fail when an LD is deleted during reconstruction (SCGCQ00821693) - Fixed an issue where firmware would go to MonTask when making VDs offline while a Rebuild is occuring under IO (SCGCQ00825775) - Fixed an issue where commands to FDE (SED) drives may not respond within 500ms (SCGCQ00844445) - Fixed an issue where the firmware may allow the OS to change VD cache settings (SCGCQ00844447) - Fixed an issue where HDD/SSD microcode updates may report fail even though they succeeded (SCGCQ00553252) - Fixed an issue where slot numbers may be wrong systems with 32 drive slots (SCGCQ00529592)
Features and Enhancements - Improved Jetstress / Exchange latencies (SCGCQ00717161) - Added support for Learn Cycle debug commands in firmware (SCGCQ00814991) - Added ability to clear the MaintainPdFailHistory (SCGCQ00474559) - Improved drive rebuild performance (SCGCQ00427581) - Improved Cache Offload error handling on read from the flash interface (SCGCQ00524419) - Improved average latency and max latency in Windows operating systems (SCGCQ00671670) - Implemented power throttling (SCGCQ00457280) - Improved load balancing on >2 drive R1 VD of SSDs (SCGCQ00593588) - Added ability for firmware to report the temperature for all SAS/SATA HDD/SSD (SCGCQ00490822) - Added support for communicating with multiple controllers over I2C (SCGCQ00485924) - Added a retry of IO commands as LDIO if raid map sync is failed with MFI_STAT_CONFIG_SEQ_MISMATCH (SCGCQ00651680) - Added support for Self-Encrypting Drives in JBOD mode (SCGCQ00399739) - Improved accuracy of ROC temperature display (SCGCQ00575037) - Increased range of priorities for background tasks, CC/PR/Rebuild/Migrate/BGI (SCGCQ00313882) - Added support for 512e drives (SCGCQ00619363) - Prevented a situation where there may be excessive overtemp warnings in the firmware log (SCGCQ00820067) - Improved DDR3 signaling margin during heavy I/O workloads (SCGCQ00824711)
Version 24.2.1-0051 - New Function Only/No fixes ------------------- lnvgy_fw_sraidmr_1215-24_9_0-0026_windows_32-64_chg
BIOS and Firmware Update for ServeRAID M5200 Series SAS/SATA Controllers
Adapters Supported: ServeRAID M5225-2GB SAS/SATA Controller ServeRAID M5210e SAS/SATA Controller for System x ServeRAID M5210 SAS/SATA Controller
NOTE TO SERVICE - Reference RETAIN #N/A
Version 24.9.0-0026 - High Impact/High Probability of Occurrence ------------------- Fixes: - Fixed an issue in iMR mode where "Controller encountered fatal error and was reset" errors can occur under heavy IOs and background operations (SCGCQ00894643) (RETAIN Tip H001154) - Fixed an issue where driver health protocol may report unhealthy if more than 64 VDs are configured (SCGCQ00830918) - Fixed an issue where VDs beyond 64 may not be reported in legacy OpROM pOST display list (SCGCQ00826174) - Fixed an issue where "StorCLI /cX add VDs detailed" menu provides incorrect syntax for R1E (SCGCQ00501437) - Fixed an issue where CLI "show all" command fails with error "Error! App exec failed in FW PCLIError! Internal" (SCGCQ00518543) - Fixed an issue where the dedicated hot spare option in HII may be grayed out for RAID1 under the profile based configuration menu (SCGCQ00873973) - Fixed an issue where HII drive management and enclosure management may report only port 4-7 connected when ports 0-3 and 4-7 are connected to the same expander (SCGCQ00870398) - Fixed an issue where HII may hang or multiple enclosure entries will be seen in enclosure eanagement form after repeated navigation (SCGCQ00878350) - Fixed an issue where a 15-character VD name is truncated to 14 characters (SCGCQ00854606) - Fixed an issue where profile-based VD creation menu doesn't explain each profile's drive selection criteria (SCGCQ00884229) - Fixed an issue where some overlap may be seen during legacy OpROM pOST when more than one controller is present and there are boot messages for the second controller (SCGCQ00866982) - Fixed an issue where MSM may display "The requested command has invalid arguments" when 'Schedule Consistency Check' is selected (SCGCQ00832908) - Fixed an issue where a controller may not be accessible through RAIDLink until the OS is up and running (SCGCQ00847001) - Fixed an issue where drive firmware on certain SED/FDE SATA SSDs could not be updated (SCGCQ00856977) - Fixed an issue where an error "MonTask: line 1272 in file ../../raid/1078dma.c" may occur while running IO in a virtual machine (SCGCQ00822403) - Fixed an issue where a JBOD PD may not be able to be converted to Unconfigured Good in Solaris (SCGCQ00834552) - Fixed an issue where an "Instruction TLB" error may occur when a StorCLI /cX set autopdcache=r0 or similar command is executed (SCGCQ00835586) - Fixed an issue where firmware may hang in multipath scenario when cable pull/push test is performed at the same time as an array property change is issued (SCGCQ00836496) - Fixed an issue where a secure erase on a SATA SSD in HII may fail or lead to a hang in HII (SCGCQ00841317) - Fixed an issue where the controller may reset in some situations with media errors and VD undergoing manual rebuild after drives are taken offline, under stress (SCGCQ00825126) - Fixed an issue where the 3x08 chip Errata ID - 2012-RC714 may lead to PCIE 3.0 hot reset non-compliance (SCGCQ00829129) - Fixed an issue where rebuild performance on RAID1 is always slower than other RAID configurations (SCGCQ00841359) - Fixed an issue where patrol read properties window won't show up in MSM (SCGCQ00797676) - Fixed an issue where an L2 unrecoverable cache error may be seen during patrol read properties modification when VDs are made offline (SCGCQ00847162) - Fixed an issue where a VD erase may not resume after reboot when a CacheCade VD is present (SCGCQ00851810) - Fixed an issue where controller boots to OS even when pinned cache is present and boot error handling is set to stop on error (SCGCQ00860503) - Fixed an issue where firmware cannot detect some newly-inserted drives (SCGCQ00860270) - Fixed an issue where "MonTask: line 1518 in file ../../raid/2108vI2o.c" may occur in conjunction with a VMWare Guest abrupt shutdown (SCGCQ00852670) - Fixed an issue where connector name field is displayed as port 4-7 when actually both ports 0-3 and 4-7 are connected (SCGCQ00868215) - Fixed an issue where "line 243 in file ../../raid/1078int.c" error may occur during rebuild (SCGCQ00856256) - Fixed an issue where no error message shown when there is a failure to disable SSD caching of an offline VD (SCGCQ00905310) - Fixed an issue where "1 JBOD(s) found on the host adapter" may appear in legacy OpROM post even though no JBODs are present (SCGCQ00882729) - Fixed an issue where firmware may halt with "line 245 in file ../../raid/1078int.c" in log under stress with maximum VDs and background operations (SCGCQ00890133) - Fixed an issue where the HII may report a different Battery/SuperCap status than MSM/StorCLI would (SCGCQ00879726) - Fixed an issue where the firmware may hang due to an "Illegal Instruction Exception" error (SCGCQ00850441) - Fixed an issue with less-than-expected performance of Write-Verify SCSI commands in RAID0 (SCGCQ00866261)
Enhancements: - Added more descriptive port information to include link width in PHYs (SCGCQ00854621)
Version 24.7.0-0056 - Moderate Impact/Low Probability of Occurrence ------------------- Fixes: - Fixed issue where UNMAP or TRIM commands might fail on SSD in JBOD Mode (SCGCQ00853147) (RETAIN Tip H214381) - Fixed an issue where an iMR RAID configuration may fail to import to 12gbps MR (SCGCQ00833765)
Version 24.7.0-0052 - High Impact/High Probability of Occurrence ------------------- Fixes: - Incorporated MegaRAID FW stack 6.6 - Changed Data Protection default from “Enabled” to “Disabled” when configuring a virtual-drive through HII - Fixed an issue where a drive reporting itself as 0MB shows up as unconfigured good (SCGCQ00784319) - Fixed an issue where the firmware might hang during error handling of sense code 0b/47/xx (SCGCQ00701744) - Fixed an issue where some SSDs may cause the system to hang at POST (SCGCQ00595506) - Fixed an issue where R0, R5 and R6 read bandwidth dropped abruptly at 256K block size with 24drive SATA SSD configuration (SCGCQ00609808) - Fixed an issue where RAID10 performance is under-performing (SCGCQ00794773) - Fixed an issue where in some cases the LSI chip would convert an incoming chip reset to a system reset (SCGCQ00664261) - Fixed an issue where many pRdm/pRequest/pReply messages would appear in LSI firmware log (SCGCQ00752195) - Fixed an issue where when a consistency check is aborted, the adapter may reset (SCGCQ00819811) - Fixed an issue where Patrol read does not start and gives incorrect warning message (SCGCQ00728404) - Fixed an issue where the controller may reset unexpectedly during error handling (SCGCQ00825126) - Fixed an issue where StorCLI may report Enclosure Not Found (SCGCQ00843146) - Fixed an issue where x880 nodes may report excessive 'SCL STUCK Low! While I2C Rx Receiving Data' in LSI firmware log (SCGCQ00728935) (Retain TIP H213891) - Fixed an issue where the command timeout for FED (SED) drives is too short, resulting in 05/2C codes (SCGCQ00797355) - Fixed an issue where bandwidth falls as queue depth increases on R5/R6 24/40 drive SAS HDD configurations at 256k block size (SCGCQ00701768) - Fixed an issue where some drives may get bad PI CRC even though data is correct resulting in drive failure (SCGCQ00819314) - Fixed an issue where the firmware may exposed more than 240VDs to host even if host driver doesn't support that many (SCGCQ00829214) - Fixed an issue where MSM may display an error message if 'Schedule Consistency Check' is selected (SCGCQ00832908) - Blocked CacheCade deletion when multiple CacheCade volumes are associated with Virtual Drives, preventing data issues (SCGCQ00838953) (Retain TIP H214323) - Fixed an issue where the MegaRAID POST may display "The native configuration is not supported by current firmware" (SCGCQ00812842) - Fixed an issue where the EFI Driver Health Protocol will show Fail when more than 64VDs are configured (SCGCQ00830918) - Fixed an issue where OS may BSOD when upgrading firmware on a controller the OS has booted from if Online Controller Reset is performed (SCGCQ00825725) - Fixed an issue where controller wouldn't go to halt state after multi-bit ECC errors detected (SCGCQ00783186) - Fixed an issue where Opal-compliant FDE (SED) drives were being marked as FDE/SED when they shouldn't be (SCGCQ00788250) - Fixed an issue where correctable ECC error interrupts were being masked (SCGCQ00814997) - Fixed an issue where reconstructed VDs were not being reflected in MSM until system reboot (SCGCQ00839018) - Fixed an issue where controller commands may fail when an LD is deleted during reconstruction (SCGCQ00821693) - Fixed an issue where firmware would go to MonTask when making VDs offline while a Rebuild is occurring under IO (SCGCQ00825775) - Fixed an issue where commands to FDE (SED) drives may not respond within 500ms (SCGCQ00844445) - Fixed an issue where HDD/SSD microcode updates may report fail even though they succeeded (SCGCQ00553252) - Fixed an issue where slot numbers may be wrong systems with 32 drive slots (SCGCQ00529592)
Features and Enhancements - Improved Jetstress / Exchange latencies (SCGCQ00717161) - Added support for Learn Cycle debug commands in firmware (SCGCQ00814991) - Added ability to clear the MaintainPdFailHistory (SCGCQ00474559) - Improved drive rebuild performance (SCGCQ00427581) - Improved Cache Offload error handling on read from the flash interface (SCGCQ00524419) - Improved average latency and max latency in Windows operating systems (SCGCQ00671670) - Implemented power throttling (SCGCQ00457280) - Improved load balancing on >2 drive R1 VD of SSDs (SCGCQ00593588) - Added ability for firmware to report the temperature for all SAS/SATA HDD/SSD (SCGCQ00490822) - Added support for communicating with multiple controllers over I2C (SCGCQ00485924) - Added a retry of IO commands as LDIO if raid map sync is failed with MFI_STAT_CONFIG_SEQ_MISMATCH (SCGCQ00651680) - Added support for Self-Encrypting Drives in JBOD mode (SCGCQ00399739) - Improved accuracy of ROC temperature display (SCGCQ00575037) - Increased range of priorities for background tasks, CC/PR/Rebuild/Migrate/BGI (SCGCQ00313882) - Added support for 512e drives (SCGCQ00619363) - Prevented a situation where there may be excessive over temperature warnings in the firmware log (SCGCQ00820067) - Improved DDR3 signaling margin during heavy I/O workloads (SCGCQ00824711)
Version 24.2.1-0052 - New Function Only/No fixes ------------------- lnvgy_fw_sraidmr_5200-24_9_0-0026_windows_32-64_chg
=============================================================================== Emulex iSCSI Device Driver for Windows ===============================================================================
Device Driver Version: 10.6.218.0 Supported On: LLenovo System x Rack and Flex
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 - iSCSI driver's usage of EQ/CQ/RQ doorbell formats for XE-102 and XE-104 - Mis-match in product version between NIC and iSCSI drivers - Lun Reset/Target Reset can hang under some errors in Windows iSCSI driver - Random crashes while load/unload surf driver
Device Driver Version: 10.0.732.0 Supported On: IBM System x
Problems Fixed: - Windows2012 blue screen dpc_watchdog_violation 256 targets during driver disable/enable - Gracefully Handle insufficient buffer ioctl response - Support logging into as many iSCSI targets as the firmware will report - Remove link speed on properties page iSCSI driver - DVL log creation fails for 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
* Add feature: Generate firmware core dump for Hilda device, and attach it to kernel memory dump when system bug checked.
version 5.3.23.0403 and 5.3.22.0317
* [ER116079] Zero minimum and maximum bandwidth % is allowed on partitioned function. Resolution: Changed minimum bandwidth % to 1.
* [ER112013] Incorrect error message while adding partitioned to LACP team. Resolution: Added new error message.
* [ER131945] VF driver crashes on installation. Resolution: Added checks to avoid accessing bad memory.
version 5.3.22.0317 and 5.3.20.0208
* [ER114642] [MSFT Bug] Team creation failure on Win2k12R2 Resolution: Added Microsoft suggested solution to get correct interface index.
* [ER111036] BSOD on launching QCCCLI after induced firmware hang and health monitoring is disabled. Resolution: Corrected handling of custom partitioning related OIDs.
version 5.3.20.0208 and 5.3.19.0130
* [ER113416] LRO stops working after changing MTU multiple times while traffic is running. Resolution: Corrected calculation of LRO session count.
* [ER113229] VLAN addition fails on team after selecting function 1 as preferred primary. Resolution: Corrected bind parameters used to create virtual.
* [ER115371] Personality change fails through NPM tab for a single port card. Resolution: Unwanted validation removed.
version 5.3.19.0130 and 5.3.16.0107
* [ER113736] QLogic management service log file is not getting cleaned from system. Resolution: Removed log file generation for release build.
* [ER114734] Ping between VMs not working when vSwitch created from 2 SILB teams are assigned to VM. Resolution: Fixed OID return processing.
* [ER112407] VM-VM traffic on VMNIC over teams with NPAR Resolution: Set learned MAC filters correctly upon switch over.
* [ER114660] Wrong PCI information shown in Teaming Tab Resolution: Corrected bus speed display.
* [ER113992] NPAR tab fails to show information. Resolution: Used locale independent string operations.
* [ER114086] Memory leak in mac learning service application. Resolution: Resource are released when not in use.
* [ER113933] Intermediate ping drop seen while running e-switch ping Resolution: Corrected VLAN id learning.
version 5.3.12.0925 and 5.3.11.0820
* [ER109812] LSO V1 parameter name is not clearly visible on W2K12 OS Resolution: Reduced string length to make LSO V1 parameter name clearly visible
* [ER111401] SRIOV: Windows VF - PVID functionality not working on Hilda Resolution: Fixed inline VLAN tagging and MAC filter programming for VF - PVID functionality
* [ER109211] Multitenancy feature support. Resolution: The following behaviour has changed from the previous release: 1. Encapsulated ICMP and UDP packets are now gracefully failed if stack requests an offload. (This change would stay until we hear back from Microsoft on bugs: 486232, 483965) 2. By default Encapsulation offload will be disabled on Win2012. (This change will stay until Microsoft resolves HCK bugs: 488699, 488682) 3. By default Encapsulation offload would not be supported on Win2012 R2. (This change will stay until Microsoft resolves all Win2012 R2 issues) 4. Encapsulation offload is now marked as Not Supported if NPAR is in configuration.
* [ER109865] SRIOV - PF reporting error on request from VF for configuring guest vlan after firmware reset. Resolution: As part of reset recovery, do not query the ACL and re-configure the VLAN
* [ER110583] Driver fail to load when load unload is run on NIC interfaces and FCOE driver is loaded. Resolution: Made 8300 (Hilda) specific changes to handle cases where FCoE driver is participating in IDC during NIC driver unload.
* [ER108280] Driver should make way for a new dump to be created once the user has retrieved a previously created dump. Resolution: Cleared state flag after retrieving the minidump
version 5.3.11.0820 and 5.3.10.0808
* [ER110279] LSO statistics don’t get updated for NVGRE traffic. Resolution: Corrected LSO counters increment for NVGRE traffic.
* [ER110302] E-switch traffic doesn't work for legacy VMNICs on W2K12-R2 HyperV. Resolution: MAC learning service has been updated to use correct WMI classes for WS2012-R2.
* [ER110369] System crash in HyperV environment when OS doesn’t allocate adequate number of MSIX interrupts. Resolution: Corrected handling of insufficient MSIX interrupts condition in the driver.
* [ER110337] Soft reset failures results in driver unload intermittently. Resolution: Corrected soft reset recovery handling in the driver.
version 5.3.10.0808 and 5.3.9.0801
* [ER109611]: QInstaller entry persists in Control Panel even after uninstallation of driver installed through Mini-installer. Resolution: Fixed bug in code which prevented the QInstaller entry from being cleaned up.
* [ER109937]: Buffer overrun in teaming configuration dll. Resolution: Added validation to avoid buffer overrun.
* [ER109211]: Multitenancy egress support not complete. Resolution: Added support for stateless offloads on NVGRE packets.
version 5.3.9.0801 and 5.3.8.0718
* [ER108978]: Less number of VMQs get created for an exported NIC function assigned to 1 VM and 8 synthetic vNIC. Resolution: Report correct max number of VMQs to NDIS.
* [ER108772]: MMC hangs when NIC Partitioning tab is opened from two interfaces. Resolution: Removed deadlock on reading flash from multiple pages.
* [ER109633]: System BSOD while adding/removing teaming driver. Resolution: Corrected memory copy in driver which was going beyond page boundary.
* [ER108547]: Adapter reset when NIC+iSCSI drivers are reverted back to MSIX interrupt mode from legacy. Resolution: Fixed control patch command and added marking/unmasking while using legacy mode interrupt.
version 5.3.8.0718 and 5.3.7.0714
* [ER106100]: QLE8262: Data corruption seen while running Sanity test on 8 ports. Resolution: Removed Tx health monitoring mechanism till the delayed DPC scheduling issue is resolved by Microsoft.
* [ER107254]: MSIx resource issue. Driver gets yellow banged. Resolution: Handled resource list properly while moving resources away. The variable-length array inside the structure is kept in accordance with the newly reduced number of resources in the list.
* [ER109211]: Multitenancy Egress support. Resolution: Added Egress support for Multitenancy.
version 5.3.7.0714 and 5.3.6.0707
* [ER108676]: Large VMQ Number Exceeded IntrSrc/IntrObj. Resolution: Added interrupt source availability check.
* [ER104913]: NIC driver unloads after resetting. Resolution: Added lagged DPC execution detection and skipped the mailbox hung check accordingly.
* [ER108486]: Teaming driver crashed Resolution: Add the boundary check and stop trashing adjacent memory
* [ER108351]: Link speed incorrect format when max bandwidth is set less than 10% Resolution: Used floating format specifier with 1 digit precision.
version 5.3.6.0707 and 5.3.5.0610.
* [ER106058]: NIC adapter's information tab displaying incorrect data. Resolution: Fixed inaccurate OID codes.
* [ER103090]: Added support to get card temperature. Resolution: Card temperature is displayed on device property page.
version 5.3.5.0610 and 5.1.18.0312.
* [ER106790]: Windows Server 2012 Hyper-V (x64) experiences a BSOD if NIC function enabled in the server profile. Resolution: Corrected an illegal write access to BAR0 Offset 0.
* [ER107161]: Low throughput and high number of pause-frame with network stress on VMQ in HyperV environment on WS2012. Resolution: Low throughput issue has been addressed by correctly handling of VMQ traffic interrupt masking/unmasking in the WS2012 driver.
* [ER105667]: Quinlan IPv6 I/O very slow when TCP Offload Checksum IPv6 is disabled. Resolution: Disabled LRO when IPv6 checksum is disabled.
version 5.1.18.0312
* Added support for 8300-series controllers with SR-IOV support for Microsoft Windows Server 2012 HyperV, and Windows Server 2012 VF driver for VMWare ESX.
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 assistance, contact QLogic Technical Support at: http://support.qlogic.com
Trademarks
Accelera, Accelerating Cluster Performance, InfiniCon Systems, InfiniNIC, InfiniPath, InfiniView, QLA, QLogic, the QLogic logo, NetXen, the NetXen logo, Intelligent NIC, 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 2014. All rights reserved worldwide. QLogic, the QLogic logo, and the Powered by QLogic logo are registered trademarks of QLogic Corporation. All other brand and product names are trademarks or registered trademarks of their respective owners. qlgc-lnvgy_dd_cna_5_3_29_801_windows_32-64_chg
=============================================================================== Emulex 16GB FC Firmware Package ===============================================================================
Firmware Version: 10.6.228.14 Supported On: Lenovo System x Rack and Flex
Problems Fixed: - None
Incremental Interoperability: - This software/firmware is only supported with Lenovo Branded System BIOS - xClarity firmware download on specific platforms (requires minimum IMM version-please consult Lenovo on the co-requisite)
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: 10.4.303.3 Supported On: System x Rack
Problems Fixed: - VF speed is showing wrongly in ethtool and Ocm. - Firmware download using ethtool -f command is failing with opcode 172-1 failed:status 100-208 - Can not set the DelayDeviceDiscovery to Max (255) by ASU tool - Ports intermittently logging in to Brocade switch as G-Port
Incremental Interoperability: - xHMC event notifications on specific platforms (requires minimum IMM version-please consult Lenovo on the co-requisite) - Lenovo branded versions of formerly IBM branded products
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)
Problems Fixed: - Blue screen after firmware flash to with virtualized WWPN
Incremental Interoperability: - xHMC event notifications on specific platforms (requires minimum IMM version-please consult Lenovo on the co-requisite) - Lenovo branded versions of formerly IBM branded products
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)
Incremental Interoperability: - Grantley based ITE
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: 10.2.377.18 Supported On: System x
Problems Fixed: - Link does not recover during IO + Port Cycling with FW 1.1.43.3 - BIOS versions return multiple values - LPe16k point to point (direct attached) connection not bootable in legacy BIOS - LPe16k point to point (direct attached) connection performance issues in legacy BIOS
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: 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 - Reported firwmare version in CMM is intermittently incorrect - Unable to set pause Flow control when connected to DCBX switch with PFC off - during cable pull testing a 40 second delay in link up experienced with Brocade switch ports - I/O completion and abort completion WQE are generated out of order - Unexpected WCQE in RQ CQ seen in RXRI mode while performing cable pulls
Incremental Interoperability: - Inventory support for the Flex xHMC (Chassis Hardware Management Console) - Crosslink & Express Lane on LPe16k based solutions
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
(C) Copyright International Business Machines Corporation 2008 - 2014. All rights reserved. US Government Users Restricted Rights - Use, duplication, or disclosure restricted by GSA ADP Schedule Contract with IBM Corporation.
************************************************************************** Lenovo SAS/SATA HDD/SSD Update Program Version 1.18.02 - Update Utility release for Lenovo x86 servers.
- Re-release of 1.18.01 since BJ pushed internally built 1.18.01 to OSS already.
************************************************************************** Lenovo SAS/SATA HDD/SSD Update Program Version 1.18.01 - Update Utility release for Lenovo x86 servers.
- fdrvwl.exe update program fixed to correctly call storelibir-3.dll for windows. Addresses defect ECR247638.
SATA 3.5"
- WD1004FBYZ-23YC, WD2004FBYZ-23YC to WH02 - improved error recovery for data errors.
- Other drive codes were added for support identification only.
************************************************************************** Lenovo SAS/SATA HDD/SSD Update Program Version 1.18.00 - Update Utility release for Lenovo x86 servers.
Note: * - Fixed extremely low probable data compare error on I/O held in standby when followed by power cycle; extremely low probable data compare error due to nonvolatile cache(NVC) write data not getting written to disc. RETAIN TIP: H214784 - MIGR-5098277
This package acknowledges support and may include updates for drives with product ID strings of:
SAS 3.5"
- ST2000NM0034, ST4000NM0034, ST6000NM0034 to BC6B - similar to Note * above.
- ST2000NM0054, ST4000NM0054, ST6000NM0054 to FC6B - similar to Note * above.
- ST1000NM0023, ST2000NM0023, ST3000NM0023, ST4000NM0023 to BC5J/LC5G - Enhance Dither Algorithm to prevent dwelling during Idle environments. - ST2000NM0043, ST4000NM0043 to EC5E/GC5C - Enhance Dither Algorithm to prevent dwelling during Idle environments.
SAS 2.5"
- ST300MX0012, ST600MX0012 to BH58 - similar to Note * above.
- ST300MP0035, ST600MP0035 to B686 - similar to Note * above.
- ST300MP0045, ST600MP0045 to E688 - similar to Note * above.
- ST1000NX0313, ST2000NX0253 to LE33 - similar to Note * above.
- AL13SXB30EN, AL13SXB60EN to TC41 - initial drive release.
************************************************************************** Lenovo SAS/SATA HDD/SSD Update Program Version 1.17.00 - Update Utility release for Lenovo x86 servers.
Note: * - Fixed extremely low probable data compare error on I/O held in standby when followed by power cycle; extremely low probable data compare error due to nonvolatile cache(NVC) write data not getting written to disc. RETAIN TIP: H214784 - MIGR-5098277
Note: # - Fixed a block misalignment during a sequential read. RETAIN TIP: H214635 MIGR-5098037
This package acknowledges support of qualified devices and may include updates for drives with product ID strings of:
SAS 3.5"
- HUS724020ALS64, HUS724030ALS64, HUS724040ALS64 to K3LB - follow-on release.
- HUH728080AL520 at K6P1 - first introduction.
- HUS724020ALS61, HUS724040ALS61 to N3L6/P3LB - internal updates.
- ST3300657SS, ST3600657SS to LA5A - follow-on release.
- ST1000NM0023, ST2000NM0023, ST3000NM0023, ST4000NM0023 to BC5H/LC5F - See Note * above.
- ST2000NM0034, ST4000NM0034, ST5000NM0034, ST6000NM0034 to BC6A - See Note * above.
- ST2000NM0054, ST4000NM0054, ST6000NM0054 to EC66/FC6A - see Note * above.
- ST2000NM0043, ST4000NM0043 to EC5D/FC5B - See Note * above.
- Other drive codes are present for support identification.
SAS 2.5"
- AL14SE180E to TL40 - new release & ship
- AL14SE030N, AL14SE060N, AL14SE090N, AL14SE120N to TB41 - new release & ship
- AL13SXB30EN, AL13SXB60EN to TC40 - new release & ship
- WD3001BKHG-23D, WD6001BKHG-23D, WD9001BKHG-23D to ZA34 - follow-on release. - WD3002BKTG-23E, WD6002BKTG-23E, WD9002BKTG-23E to ZC33 - follow-on release.
- MK3001GRDB, MK6001GRDB, MK9001GRDB to TB37 - follow-on release.
- HUC109030CSS60, HUC109060CSS60, HUC109090CSS60 to J2EH/K2EJ - follow-on release. - HUC109030CSS61E, HUC109060CSS61E, HUC109090CSS61E to P2EJ - follow-on release.
- HUC101830CSS20, HUC101860CSS20, HUC101890CSS20, HUC101812CSS20 to K2H4 - follow-on release. - HUC101830CSS20E, HUC101860CSS20E, HUC101890CSS20E, HUC101812CSS20E to P2H4- follow-on release.
- HUC156030CSS20, HUC156060CSS20 to J2G6/K2GA - re-introduction. - HUC156030CSS20E to P2GA - re-introduction.
- HUC156060CSS20E to N2G6 - re-introduction.
- HUC156030CS420, HUC156060CS420 to J5G6/K2GA - re-introduction.
- HUC1012120CSS60 to K2MC - follow-on release. - HUC1012120CSS61 to P2MC - follow-on release.
- ST9146853SS to L63L - follow-on release.
- ST300MM0018, ST900MM0178 to F583 - see Note * above.
- ST1000NX0453 TO LD42 - see Note * above.
- ST300MM0008, ST900MM0168 TO L583 - see Note * above.
- ST300MX0012, ST600MX0012 to BH57 - see Note * above.
- ST300MP0045, ST600MP0045 to E687 - see Note * above. - ST300MP0035, ST600MP0035 to B685 - see Note * above.
- ST300MP0025, ST6 00MP0025 to E671/F674 - see Note * above.
- ST300MP0005, ST600MP0005 to L679/B675 - see Note * above.
- TXA2D20200GA6I, STXA2D20400GA6I, TXA2D20800GA6I, TXA2E21600GA6I to GB7P/HB7P - enhancements.
************************************************************************** Lenovo SAS/SATA HDD/SSD Update Program Version 1.16.05 - Update Utility release for Lenovo x86 servers.
This package acknowledges support and may include updates for drives with product ID strings of:
SAS 3.5"
- ESXSST1000NM0001, ESXSST2000NM0001 to BC4B - ensures that after servicing an 03/11/00 error, an index for Parity Creations is calculated correctly after a power cycle, that would otherwise make the drive unresponsive.
- ESXSHUS724020ALS64, ESXSHUS724030ALS64, ESXSHUS724040ALS64 to J3L6 - Fixed unexpected 02/04/11 error; Fixed a drive logic self-initiated reset on read/write long commands; Fixed a problem with large write sequential transfers.
SAS 2.5"
- ESXSST9500620SS, ESXSST91000640SS to BD2K - ensures that after servicing an 03/11/00 error, an index for Parity Creations is calculated correctly after a power cycle, that would otherwise make the drive unresponsive.
SATA 2.5"
- SSDSC2BB012T4I, SSDSC2BB016T4I to G201NC32 - SCT WriteSame fix.
************************************************************************** Lenovo SAS/SATA HDD/SSD Update Program Version 1.16.04 - Update Utility release for Lenovo System x servers.
This package acknowledges support and may include updates for drives with product ID strings of:
NOTE: Check notation under PCIe NVMe devices.
SATA 2.5"
- MTFDDAK120MBB, MTFDDAK240MBB, MTFDDAK480MBB, MTFDDAK800MBB to MB42 - > A drop in host power supply may cause bad block growth > The bad block table may not properly increment > Background activity may continue after STANDBY, STANDBY IMMEDIATE, or SLEEP > ID173/202 may show incorrect value when the NAND erase count exceeds 3000 > Changed DRAM timing > Event log may be programmed when supply voltage from host is out of specification
- SSDSC2BB012I4I, SSDSC2BB016T4I to G201NC32 - SCT WriteSame fix that corrects a condition that may cause the drive to not respond.
PCIe NVMe 2.5"
- SSDPE2ME400G4I, SSDPE2ME800G4I, SSDPE2ME016G4I, SSDPE2ME020G4I to 8DV1ND34 - will require two update runs and system reboots for device fw update to complete--Please See the limitations in readme for details.
> Performance improvement on sub-4KB sequential write > Fix drive reports higher than actual data units read and write > Fix drive incorrectly counts unsafe power shutdown events on warm reboots
************************************************************************** Lenovo SAS/SATA HDD/SSD Update Program Version 1.16.03 - Update Utility release for Lenovo System x servers.
This package acknowledges support and may include updates for drives with product ID strings of:
NOTE: Check notation under PCIe NVMe devices.
SAS 3.5"
- ESXSST2000NM0034, ESXSST4000NM0034, ESXSST6000NM0034 to BC67 - GA code. - ESXSST2000NM0054, ESXSST4000NM0054, ESXSST6000NM0054 to EC66 - GA code.
- ESXSWD1001FYYG-23S, ESXSWD2001FYYG-23S, ESXSWD3001FYYG-23S, ESXSWD4001FYYG-23S to XA39 - issues fixed: slow performance with PI enabled; drive hang on uncorrectable rd error in reallocation track; PERSISTANT RESERVATION command not working correctly; reliability improvements around off-track write protection.
SAS 2.5"
- ESXSTXA2D20200GA6, ESXSTXA2D20400GA6, ESXSTXA2D20800GA6, ESXSTXA2E21600GA6 to GB7P - fixes implemented to support PI issues.
- ESXSST300MP0005 to B675 - GA code. - ESXSST300MP0025 to E671 - GA code.
- ESXSST300MP0035, ESXSST600MP0035 to B683 - GA code. - ESXSST300MP0045, ESXSST300MP0045 to E685 - GA code.
- ESXSST300MX0012, ESXSST600MX0052 to BH53 - GA code.
- ESXSWD3001BKHG-23D, ESXSWD6001BKHG-23D, ESXSWD9001BKHG-23DS to ZA34 - issues fixed: slow performance with PI enabled; drive hang on uncorrectable rd error in reallocation track; PERSISTANT RESERVATION command not working correctly; drive was reporting 0B/11 instead of 03/11. - ESXSWD3002BKTG-23E, ESXSWD6002BKTG-23E, ESXSWD9002BKTG-23E to ZC33 - issues fixed: slow performance with PI enabled; drive hang on uncorrectable rd error in reallocation track; PERSISTANT RESERVATION command not working correctly; reliability improvements around off-track write protection.
SATA 3.5"
- MG04ACA200E, MG04ACA300E, MG04ACA400E, MG04ACA500E to SJ1C - GA code
SATA 2.5"
- ST9250610NS, ST9500620NS, ST91000640NS to BE2B - Update configuration change.
- MTFDDAK120MBB, MTFDDAK240MBB, MTFDDAK480MBB, MTFDDAK800MBB to MB34 - > Some Out of Range LBAs accepted by NCQ commands where they should be aborted. > SataDownShift.srt failed, Smart attribute Raw lost with perform dirty pwr cycle . > Low Power Entry Bug Identified from Marvell Code Review > Track Timeout, Marvel 373 sequencer issue > Unreasonable ECC count in SMART ID 195
************************************************************************** Lenovo Online SAS/SATA HDD/SSD Update Program for Linux. Version 1.16.00 Build ID: SAS11600
******************************************* Lenovo SAS/SATA HDD/SSD Update Program Version 1.16.00 - First Update Utility release for Lenovo System x servers.
This package acknowledges support and may include updates for drives with product ID strings of:
NOTE: Check notation under PCIe NVMe devices.
SAS 3.5"
- ESXSST1000NM0023, ESXSST2000NM0023, ESXSST3000NM0023, ESXSST4000NM0023 to BC5E
- ESXSST2000NM0043, ESXSST3000NM0043, ESXSST4000NM0043, ESXSST1000NM0043 to EC5A
- ESXSST3300557SS, ESXSST3450757SS, ESXSST3600957SS to EA54
- ESXSST3300657SS, ESXSST3450857SS, ESXSST3600057SS to BA5A
- ESXSST33000650SS to BC3E
- ESXSST1000NM0001, ESXSST2000NM0001 to BC4A
- ESXSHUS724020ALS64, ESXSHUS724030ALS64, ESXSHUS724040ALS64 to J3L5
- ESXSHUS724020ALS61, ESXSHUS724040ALS61 to N3L5
- ESXSHUS723020ALS64, ESXSHUS723030ALS64 to J3K8
- ESXSWD1001FYYG-23S, ESXSWD2001FYYG-23S, ESXSWD3001FYYG-23S, ESXSWD4001FYYG-23S to
- ESXSWD1001FYYG-23S, ESXSWD2001FYYG-23S, ESXSWD3001FYYG-23S, ESXSWD4001FYYG-23S to XA38
SAS 2.5"
- ESXSTXA2D20200GA6I, ESXSTXA2D20400GA6I, ESXSTXA2D20800GA6I, ESXSTXA2E21600GA6I to GB7H
- ESXSST300MM0006, ESXSST600MM0006, ESXSST900MM0006 to B56M; - ESXSST300MM0026, ESXSST600MM0026, ESXSST900MM0026 to E56D
- ESXST1200MM0007 to B57B - ESXST1200MM0017 to E576
- ESXSST9300605SS, ESXSST9600205SS, ESXSST9900805SS to B55E - ESXSST9300505SS, ESXSST9600105SS, ESXSST9900705SS to E558
- ESXSST9146853SS, ESXSST9300653SS to B63K
- ESXSST9146753SS, ESXSST9300553SS to E638 - ESXSST9500620SS, ESXSST91000640SS to BD2J
- ESXSHUC101212CSS60 to J2MA - ESXSHUC101212CSS61 to N2M9
- ESXSHUC156030CS420, ESXSHUC156060CS420 to J5G2 - ESXSHUC156030CSS20, ESXSHUC156060CSS20 to J2G6
- ESXSHUC156030CS420E, ESXSHUC156060CS420E to N5G2
- ESXSHUC156030CSS20E, ESXSHUC156060CSS20E to N2G4
- ESXSHUC101860CS420, ESXSHUC101890CS420, ESXSHUC101812CS420, ESXSHUC101818CS420 to J5H2 - ESXSHUC101860CS420E, ESXSHUC101890CS420E, ESXSHUC101812CS420E, ESXSHUC101818CS420E to N5H2
- ESXSHUC109030CSS60, ESXSHUC109060CSS60, ESXSHUC109090CSS60 to J2EG - ESXSHUC109030CSS61E, ESXSHUC109060CSS61E, ESXSHUC109090CSS61E to N2EG
- ESXSAL13SXB300N, ESXSAL13SXB600N to SC36
- ESXSMBF2300RC, ESXSMBF2600RC to SB2F
- ESXSMK1401GRRB, ESXSMK3001GRRB to SC2E - ESXSMK1401GRRR, ESXSMK3001GRRR to SF23 - ESXSMK3001GRDB, ESXSMK6001GRDB, ESXSMK9001GRDB to SB35
- ESXSWD3001BKHG-23D, ESXSWD6001BKHG-23D, ESXSWD9001BKHG-23D to ZA33 - ESXSWD3002BKTG-23E, ESXSWD6002BKTG-23E, ESXSWD9002BKTG-23E to ZC32
- EXSXHUSMM1620ASS20, EXSXHUSMM1640ASS20, EXSXHUSMM1680ASS20, EXSXHUSMM1616ASS20 to J4C6 - EXSXHUSMM1620ASS20E, EXSXHUSMM1640ASS20E, EXSXHUSMM1680ASS20E, EXSXHUSMM1616ASS20E to N4C6
SATA 3.5"
- WD1003FBYX-23Y7B0, WD2003FYYS-23W0B0 to WA35
- WD2000FYYZ-23UL, WD3000FYYZ-23UL, WD4000FYYZ-23UL to WD37 - Resolves drive initiated reset (DBS (Debug Stop event)) encountered under a narrow timing window (~1ns) for specific command/access patterns, due to the drive’s handling of command automation where it was possible for an internal hardware resource conflict to occur between two tasks.
- ST33000650NS to BB3A
- ST1000NM0011, ST2000NM0011 to BB4A
- ST1000NM0033, ST2000NM0033, ST3000NM0033, ST4000NM0033 to BB5A
- HUA723020ALA640 to MK7OA930 - HUA723030ALA640 to MKAOA930
SATA 2.5"
- ST9250610NS, ST9500620NS, ST91000640NSto BE2A
- MTFDDAK120MBB, MTFDDAK240MBB, MTFDDAK480MBB, MTFDDAK800MBB to MB33
- SSDSC2BA200G3I, SSDSC2BA400G3I, SSDSC2BA800G3I to 5DV1NA37
- MZ7GE240HMHP-000M3, MZ7GE480HMHP-000M3, MZ7GE960HMHP-000M3 to EXST0CA34
PCIe NVMe 2.5"
- SSDPE2ME400G4I, SSDPE2ME800G4I, SSDPE2ME016G4I, SSDPE2ME020G4I to 8DV1ND32 - will require a system reboot for device fw update to complete. lnvgy_fw_hdd_sas-1_18_02_windows_32-64_chg
Intel(R) Chipset Device Software for Windows
NOTE TO SERVICE - Reference RETAIN TIP #NA
Version 9.4.2.1017 ------------ - Initial release intc-lnvgy_dd_inteldp_9_4_2_1017_windows_32-64_chg
=============================================================================== Emulex 16GB FC Firmware Package ===============================================================================
Firmware Version: 10.6.228.14 Supported On: Lenovo System x Rack and Flex
Problems Fixed: - None
Incremental Interoperability: - This software/firmware is only supported with Lenovo Branded System BIOS - xClarity firmware download on specific platforms (requires minimum IMM version-please consult Lenovo on the co-requisite)
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: 10.4.303.3 Supported On: System x Rack
Problems Fixed: - VF speed is showing wrongly in ethtool and Ocm. - Firmware download using ethtool -f command is failing with opcode 172-1 failed:status 100-208 - Can not set the DelayDeviceDiscovery to Max (255) by ASU tool - Ports intermittently logging in to Brocade switch as G-Port
Incremental Interoperability: - xHMC event notifications on specific platforms (requires minimum IMM version-please consult Lenovo on the co-requisite) - Lenovo branded versions of formerly IBM branded products
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)
Problems Fixed: - Blue screen after firmware flash to with virtualized WWPN
Incremental Interoperability: - xHMC event notifications on specific platforms (requires minimum IMM version-please consult Lenovo on the co-requisite) - Lenovo branded versions of formerly IBM branded products
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)
Incremental Interoperability: - Grantley based ITE
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: 10.2.377.18 Supported On: System x
Problems Fixed: - Link does not recover during IO + Port Cycling with FW 1.1.43.3 - BIOS versions return multiple values - LPe16k point to point (direct attached) connection not bootable in legacy BIOS - LPe16k point to point (direct attached) connection performance issues in legacy BIOS
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: 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 - Reported firwmare version in CMM is intermittently incorrect - Unable to set pause Flow control when connected to DCBX switch with PFC off - during cable pull testing a 40 second delay in link up experienced with Brocade switch ports - I/O completion and abort completion WQE are generated out of order - Unexpected WCQE in RQ CQ seen in RXRI mode while performing cable pulls
Incremental Interoperability: - Inventory support for the Flex xHMC (Chassis Hardware Management Console) - Crosslink & Express Lane on LPe16k based solutions
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
Please visit www.qlogic.com for full documentation, additional drivers, and management applications.
Version 3.2.5.6 Fixes and Enhancements: - Boot FW, UEFI, HCM related fixes - Rebranding changes
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 "Lenovo 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 Lenovo 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 Lenovo 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 Lenovo 16G Mezzanine Fabric Adapter.
HBA - Fibre Channel 16G Support
Fixes and Enhancements: N/A brcd-lnvgy_fw_cna_3_2_5_6_windows_32-64_chg
Firmware Version: 10.6.228.13 Supported On: Lenovo System x Rack and Flex
Problems Fixed: - Improved link negotiation with 1G switches
Incremental Interoperability: - This software/firmware is only supported with Lenovo Branded System BIOS - UFP mode and concurrent personality are now supported (175594, 176654) - Flex System CN4052 R 2-port 10Gb Virtual Fabric Adapter - Lenovo 00AG572 VFA5 2x10GbE SFP+ PCIe Adapter - Lenovo 00AG582 VFA5 2x10GbE SFP+ PCIe FCoE/iSCSI Adapter - Lenovo 00AG562 VFA5 ML2 2x10GbE SFP+ Adapter - xClarity firmware download on specific server platforms (requires minimum IMM version-please consult Lenovo on the co-requisite) (limited to CN4052 R,00AG562,00AG572,00AG582)
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.(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)
Firmware Version: 10.4.303.14 Supported On: System x Rack
Problems Fixed: - Improved handling of ASU/UCM deployed configurations.
Incremental Interoperability: - Lenovo branded versions of formerly IBM branded products - Lenovo xHMC support for inventory management - Mixed and Concurrent Mode storage (refer to documentation at Emulex.com for more information on this functionality). - 16 total pf's per ASIC in UMC (vNIC2) Mode. Was previously limited to 8 pf's per ASIC.
Known Issues: - UFP mode and concurrent personality are not supported at this time. This will be resolved in the next release (175594, 176654)
- 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)
- 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)
- 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 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 servers.
Problems Fixed: - MAC/PHY loopback test intermittent failures - If BIOS enables "Above 4G decoding", Emulex iSCSI driver is not loaded - Error message appears when saving changes to minimum BW on run time FCoE configuration - Eighth NIC function doesn't come up in UMC mode when configured in PXESelect
Incremental Interoperability: - Lenovo branded versions of formerly IBM branded products
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)
- 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)
- 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)
- 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.
Problems Fixed: - Ethtool shows “speed=unknown� sometimes for slave interface. (vNIC & UFP) - SVID values are not present under PXE select utility in vNIC1 mode - UMC Mode/Personality change using IBM ASU tool does not reflect in OCM
Incremental Interoperability: - Grantley based ITE - IBM Flex System CN4052 2-port 10Gb Virtual Fabric Adapter - IBM Flex System CN4058S 8-port 10Gb Virtual Fabric Adapter
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)
- 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)
- 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)
- 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: 10.2.377.20 and 10.2.377.23
Supported On: System x and BladeCenter
Problems Fixed: - 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) - "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) - 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) - UE seen while booting the system after a CoreDump - When NIC traffic has run on all the 8 vNICs of an ASIC, the vnics of first port are not honoring the assigned BW - SRIOV performance & Fairness for 16VMs is low on Linux and VMware - ML2 VFA5 device name is show "N/A" under IMM web and CLI of server management>Adapters - BW changes not getting updated with vNIC1 mode - Enable 100M/10M speed in set speed for SGMII interface - UCM/vNIC2, Ports/PCI functions supporting FCoE need to be differentiated to make it easier for the user - UCM/vNIC2 ability to program VirtWWN/WWPN independent of BFS Enable - iSCSI boot not working on higher order PCI functions (PF8 11)on 4 port controllers on ARI enabled systems - Can't install Windows2012 onto the iSCSI LUN whose LUN ID is not 0 - Fixed several issues related to IBM Agentless inventory management - Under heavy bi-di traffic with XE102/XE104 network disconnects can occur during failover/switch side link bounce
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)
- 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)
- 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)
- 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: 10.2.261.41 Supported On: IBM System x
Problems Fixed: - Hit BSOD during VMQ overnight traffic test with cpu affinity change - IBM VFA5 Adapter causing software panic on IBM G8246 Switch in UFP mode - LPVIDs are not getting saved in vNIC2 mode if we change mode from vNIC1 mode to vNIC2 - In MC VINC1, VNIC2, UFP modes, user should see only 4 functions per port for 2 port card
Incremental Interoperability: - Emulex VFA5 2x10 GbE SFP+ Adapter and FCoE/iSCSI SW for IBM System x - Emulex VFA5 2x10 GbE SFP+ PCIe Adapter for IBM System x - Emulex VFA5 2x10 GbE SFP+ Integrated Adapter for IBM System x
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: 2.02x13 Boot Code Version: 5.30a7 Supported On: Lenovo System x and Flex
Problems Fixed: -
Incremental Interoperability: - This software/firmware is only supported with Lenovo Branded System BIOS
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.(159225)
Firmware Version: 2.02x11 Boot Code Version: 5.30a6 Supported On: System x, BladeCenter and Flex
Problems Fixed: - SAN boot stopped working after firmware/boot BIOS upgrade
Incremental Interoperability: - Lenovo branded versions of formerly IBM branded products
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.(159225)
Firmware Version: 2.02x11 Boot Code Version: 5.30a3 Supported On: System x, BladeCenter and Flex
Problems Fixed: - None
Incremental Interoperability: - Grantley based ITE
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.(159225)
Firmware Version: 2.02x11 Boot Code Version: 5.30a3 Supported On: System x and BladeCenter
Problems Fixed: - Roll boot code version number to avoid conflict with IBM UXSPi updates
Incremental Interoperability: - Inventory support for the System x xHMC (Chassis Hardware Management Console) for IBM Rack FC HBAs - Grantley Rack Servers
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.(159225)
Firmware Version: 2.02x11 Boot Code Version: 5.13x4 Supported On: IBM System x, BladeCenter, and Flex
Problems Fixed: - 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.(159225)
Firmware Version: 2.01a11 Boot Code Version: 5.12a13 Supported On: IBM System x and BladeCenter
Problems Fixed: - Emulex FC Adapter does not auto-enable SAN boot mode when boot target configuration is made via IFM (BOFM) - Emulex FC configuration utility is in the wrong F1 setup screen - IBM CIOv adapter connected to Brocade 8 Gig FC BladeCenter switch intermittently Links up at less than 8 Gig
Firmware Version 2.01a3 Boot Code Version: 5.12a5 Supported On: IBM System x and BladeCenter
Incremental Interoperability: - Enhanced timeout notifications for processing abort failures.
Problems Fixed: - All multi-abort I/Os are returned on the same ring from which the ABORT_MXRI command was received. - Changed the TRCV response in the abort state to CMD RJT to enable the TRSP to be processed. - Corrected an issue with blocked TSEND operations. - Corrected a buffer leak issue with RRQ exchanges. - Corrected an issue where the TRCV IOCB was de-queued improperly causing a XFR_RDY flood. - Corrected an issue where the OXID field was not loaded in the XRI_ABORTED IOCB. - Corrected an issue in recognizing whether an EEPROM is present.
Version 1.09.04 - Moderate Impact/Low Probability of Occurrence --------------- Contains: BIOS Firmware UEFI Driver 08.21.06.00 09.00.02.00 10.00.00.00 Fixes: - Fixed issue where HDD removal not properly reported as Target Device missing when attached to adapter through an expander. (SCGCQ00777858) - Fixed issue where Fault State (0x265d) is observed in driver logs while connecting an enclosure with fully populated SATA drives (SCGCQ00856085) - Fixed issue where 0x7C54 fault seen during I/T switching (SCGCQ00847013) - Fixed issue where certain drive pull scenarios result in longer time to discovery running (~3+ seconds) (SCGCQ00843759) - Improved PCIe Config operation handling to avoid 0x0D03 faults (SCGCQ00836878) - Fixed issue where controller could fault with 0x4A00 during link break (SCGCQ00793956) - Fixed the occurrence of fault 0x1708, when a free MID is attempted to be failed to the host. (SCGCQ00826420) - Fixed occurrence of fault 0x5814/0x5813 while handling Task Management during IO Timeout (SCGCQ00865024) - Fixed system hang after adapter reset with heavy IO load (SCGCQ00877099) - Fixed issue with slow BIOS interface response with certain hardware e nvironments (SCGCQ00888867) - Fixed UEFI issue where RSOD seen with SAS 12G HBA connected to an enclosure with a failed drive.(SCGCQ00843066) - Fixed UEFI issue where in an environment with multi-lun devices where lun size varies, backup GPT corruption can occur (SCGCQ00907441)
Enhancements: - In BIOS added a new field to Device Properties called "Logical Sector Size". The possible values are 512B, 4K and Unknown. (SCGCQ00887338) - Improved the response time in UEFI HII utility for SAS3 IT/IR Controllers (SCGCQ00821793)
Version 1.07 - Moderate Impact/Low Probability of Occurrence --------------- Contains: BIOS Firmware UEFI Driver 08.15.01.00 07.00.01.00 08.00.00.00
Firmware: Fixes: - Fixed issue where 0xD025 fault appeared while running non-FastPath I/O to SATA disk with outstanding FastPath I/Os to SAS disk (SCGCQ00604668) - Fixed fault 0xD071 while running stress and glitching all 4 phys as part of a cable break test. (SCGCQ00590502) - Fixed MPI Power Management Global Request Change to Reduced or Standby state results in "Operation State Not Supported" or "IOC Busy" (SCGCQ00584284) - Fixed issue where when transitioning MPI Power Management speed from lower to higher rate the command falsely reported as successful (SCGCQ00584288) - Fixed issue with mixture of FastPath IO's and standard firmware path IO's timing out on SATA disks (SCGCQ00576190) - Fixed SATA only issue where Multiple 'Write Verify' SCSI commands with zero data length leads to f/w fault (0x4203 or 0x4C31). (SCGCQ00549293) - Fixed fault 0x6230 while disabling expander phys and running stress in multi-path failover environment (SCGCQ00692652) (SCGCQ00699059) - Fixed various conditions causing 0x6230 fault when cable breaking (SCGCQ00669775) (SCGCQ00699057) (SCGCQ00643835) - Fixed issue where rapid successive controller resets would cause HBA to stop responding on the PCIe bus (SCGCQ00556766) - Fixed issue where when phy power management is enabled, TARGET RESET function fails yet command is reported a success. (SCGCQ00570820) - Fixed issue where PCIe Link Down can cause transmit margin field to not reset resulting in PCIe validation tests failing (SCGCQ00663185) - Fixed issue where SCSI command is not failed with expected response when SATA drive undergoing a format operation and reserved fields are set as Non-Zero value (SCGCQ00679566) - Fixed issue where read log ext command wasn't issued by PL FW to recover from error during the failure of write verify command (SCGCQ00675166) - Fixed IOC fault state 0x1609 caused by Message Unit Reset (SCGCQ00582515) - Fixed issue where SAS Phy page 1 and 2 error counters wrap around when maximum values are reached. (SCGCQ00651231) - Fixed fault 0x265D when running large block IO with higher Qdepth (SCGCQ00600832) - Fixed issue in Transport Layer Retries Code causing 0x265D and 0x0501 faults (SCGCQ00609576) - Fixed issue where Bad TLB entry causes 0x26B5 fault in IT/IR firmware (SCGCQ00629210) - Fixed 0x6828 fault when using large CDB (SCGCQ00614498) - Fixed fault 0x1901 seen when running Unmap IO’s during task aborts for a SATA drive (SCGCQ00711858) - Fixed issue where system gets hung & reboot while running IO's along with diag reset on the Controller which has Active and Inactive volumes. (SCGCQ00714125) - Fixed issue where cards would fault with 58B4 on power up due to incorrect SERDES setting (SCGCQ00728903) - Fixed WRITE SAME(10) and WRITE SAME(16) SCSI commands fail on last LBA of SATA Drive, incorrectly reporting logical block out of range. (SCGCQ00703433) - Fixed fault 0x4203 when WRITE AND VERIFY (16) command us executed to SATA Drive for LBA Out of Range (SCGCQ00705320) - Fixed issue where ATA Translation of sector count in sense data is inconsistent for READ DMA pass-through cmd (SCGCQ00712688) - Fixed issue where CLI command to dump L1 data cache causes 0x265D fault (SCGCQ00715317) - Fixed SATA Init problem causing 0x6230 controller fault (SCGCQ00683795) - Fixed fault 0x265D hit when booting controller with attached enclosure (SCGCQ00709712) - Fixed issue where Negotiated Physical Link Rate in SAS IO Unit Page 0 may not be correct after pulling a cable (SCGCQ00696408) - Fixed issue in NVDATA where SGPIO PHYs reversed (SCGCQ00750746) - Fixed I2C slave status interrupt not always cleared with I2C Write operation (SCGCQ00731646)
Enhancements: - Improved implementation of SCSI transport layer (SATL) handling while the SATA drive is in security 'frozen' state (SCGCQ00549302) - SCSI Transport Layer: Indicate support for the ATA Device Server Password security protocol (SCGCQ00625580) - IO Processor Target Mode performance improvements (SCGCQ00685827)
UEFI: Fixes: - Fixed issue where the navigation of HII screens was very slow when no drives are connected to an expander (SCGCQ00569123) - Reduce the boot time for UEFI driver for large configurations. Fixes a falsely reported Backup GPT corruption. (SCGCQ00498101) (RETAIN Tip H212119) - Fixed issue where HII configuration utility did not print the complete SAS address (SCGCQ00709781) - Fixed issue where View Physical Disk Properties screen on HII was not showing correct disk details (connector:encl:slot) when more than 8 disks are directly attached. (SCGCQ00691104)
Enhancements: - Obtain the Physical Slot number from the PCI Bridge first and use fall back to SMBIOS table if unable retrieve slot number from PCI Bridge. (SCGCQ00498095) - Retry Test Unit Ready command if drive foes not respond on first try. (SCGCQ00518198) - Adding a small delay (2 seconds) to get proper operation status after starting an operation (like consistency check) in HII (SCGCQ00621480) - Enhanced HII configuration utility to display tape drive information (SCGCQ00700835)
BIOS: Fixes: - Fixed issue where the system boot hangs when Legacy BIOS is disabled (using HII) on certain UEFI systems. (SCGCQ00637088) - Fixed issue where the Legacy BIOS failed to detect certain large capacity SAS drives (SCGCQ00696486)
Enhancements: - Support configurable timeout for Start Stop Unit Command. (SCGCQ00694336)
QLogic 16 Gb Fibre Channel Adapter MultiFlash Update for System x QLogic Corporation. All rights reserved.
Version 4.72.12 09/22/2015
* Flash Image Version 4.72.12
HLD47212.BIN contains: ---------------------- * FC Bootcode (2600 Series only) BIOS 3.0 version 3.29 uEFI version 6.22 signed * FC Firmware - v7.05.04 * Pegtune - v5.110.23 * CRBnit - v5.14.15
* Removed Adapter Personality field from Agentless page. (ER136194) * Microsoft signed driver.
Change History *************************************************************************************
Version 4.70.07 03/26/2015
* Flash Image Version 4.70.07
HLD47007.BIN contains: ---------------------- * FC Bootcode (2600 Series only) BIOS 3.0 version 3.29 uEFI version 6.21 Signed * FC Firmware - v7.05.04 * Pegtune - v5.110.10 * CRBnit - v5.14.15
* Fixed issue causing MD2 CLP to fail (ER135507). * Added code to fix CLP issue with 13G Dell system * Fixed issue in displaying Execution Throttle in CTRL-Q utility correctly for 4/8G HBA. * Added coded to load/start FW for OCBB/OCSD only when BIOS is disabled * Added support of Baker HBA * Added code to not configure any targets when FA Boot * LUN enabled but FA BOOT LUN not found. ER133769. * Added code to allow FA Boot LUN when FA WWPN enabled * Added options in CTRL-Q utility to enable FA WWPN and FA BOOT LUN. * Fix minor cosmetic error in display of Fcode/EFI versions for show port CLP.(ER133294) * Added support FA BOOT LUN * Added 8G support for show port clp * Changed VLAN ID max value from 4095 to 4094. * Removed spinup option from CTRL-Q. * Changed CTRL-Q to allow max value for Login Retry, * Port Down Retry, Link Down Count of 65535.ER(116099) * Removed Helga support * Added support for show port clp. * Remove EV boot order option from OEM BIOS. * Added CTRl_Q option to set VLAN_ID for FCoE HBA's(ER110593). * Changed Banner to display 'FCoE' for Schultz (ER110784). * Added additional time to login target when called after EFI driver unloads. (ER110755,ER110855) * Added code to handle boot LUN Greater Than 255 (ER109195). * Added support for Helga. * Added code to display message when 64 bit address is detected. * Added support for Execution Throttle for P3P (ER106372) * Added code to not display connection options for Hilda Trinity adapter. * Checked for invalid LUN in CTRL-Q scan devices. * Removed port number from HII formset title. This fixes a UCM parsing issue. (ER134560).
HLD32009.BIN contains: ---------------------- * FC Bootcode (2600 Series only) BIOS 3.0 version 3.24 uEFI version 6.20 * FC Firmware - v7.02.00
* QLflash version 4.1.0.5
These changes describe the changes, fixes, known issues, and release details that apply to the FC-FCoE BIOS version 3.24.
Version 3.24 * Added code to display message when 64 bit address is detected. * Added support for Execution Throttle for P3P (ER106372) * Added code to not display connection options for Hilda Trinity adapter. * Checked for invalid LUN in CTRL-Q scan devices. * Fixed bug in CTRL_Q utility that could cause some cosmetic display issues. * Added Hilda CLP _hss support. * Configured ISP24xx if on slot 0. * Added "QLogic" to PnP productid (ER104424) * Checked for 4G flag in NVRAM for Hilda FC.(ER104714)
HLD32006.BIN contains: ---------------------- * FC Bootcode (2600 Series only) BIOS 3.0 version 3.24 uEFI version 6.17 * FC Firmware - v7.02.00
* Changed Banner to display 'FCoE' for Schultz (ER110790). * Updated to get PN & Card SKU dynamically from I2C VPD for AIM (ER116141) * Updated to get PN & Card SKU dynamically from I2C VPD for AIM (ER116141) * Updated MTU size and AIM version references as per revised requirement. (ER113361) * Updated AIM config strings as per revised requirements. (ER113361) * Updated AIM support as per revised requirements (ER113361) * Added fix for boot settings persistency issue in BOFM supported legacy BIOS (ER112641) * Added support for Lenovo AIM [ER113361] * Added fix to persist user changes made on static forms [ER112639] * Added feature to dynamically list targets in HII [ER105725]
Version 3.20.03 Date: 10/01/2013 BIOS 3.0 version 3.24 uEFI version 6.10 FC Firmware - v7.01.00
* Added additional time to login target when called * after EFI driver unloads. (ER110755.ER110855) * Added code to display message when 64 bit address detected. * Added support for Execution Throttle for P3P (ER106372) * Check for invalid LUN in CTRL-Q scan devices. * Fixed bug in CTRL_Q utility that could cause some cosmetic display issues. * Configure ISP24xx if on slot 0. * Added "QLogic" to PnP productid (ER104424) * Fixed bug with CTRL-Q Loopback Utility (ER100729). * Added code to check/program Feature Config checksum. * Fix bug that could cause buffer overrun in CTRL-Q with Report Lun command (ER100294). * Added check for PCI address change when updating option card data structure. * Changed code to use 32 bit pointer to allocate/deallocate EBDA memory. * Added code to check/program Feature Config checksum. * Fixed cosmetic issue in displaying data rate in CTRL-Q * Fixed display issue with CTRL-Q Diskutility(ER98727) * Fixed Lun select issue with CTRL-Q (ER98675). * Fixed bug in displaying selectable boot lun (ER98226) * Removed 4G data rate option for OEM adapter (ER98228) * Limit leading '0's in CTRL-Q menu (ER98389). * Set 'Aggressive Cooling' flag for OCSD. * Enabled Personality Change option. * Fixed issue with not restoring defaults on port 1 (ER 97619) * Fixed issue with serdes data saved incorrectly. (ER 97654) * Added code to limit Report Lun List size to 256 luns * Added check for 0 lun count when selecting device (ER 97385) * Disable Personality Change option in CTRL-Q. * Fixed bug with WWNN handling for MCTP.(ER 96920) * Added check for Set MD2 set to 0 (restore defaults) (ER96912) * Added code to reset Port Config after Internal loopback test(FCoE). * Added support for 16 bit lun (er96090). * Add support for Set MD2 CLP string. * Added code to issue Load Flash FW (3) MB command as part ISP start sequence. (ER 94035). * Fixed code to prevent async event from causing BIOS to hang when scanning for devices (er94371). * Fix code to disable ALT+1 Hotkey (er92844) . * Fixed CTRL-Q code mailbox timeout (er92957). * Added warning message in Personality change screen. * Updated BIOS to version 3.12 of non-hilda BIOS * Removed check for Virtual Address on Cold Boot. * Changed code to check for signature aaaaaaaah before updating virtual mac address. * Changed code to ignore soft boot signature aaaaaaaah * Added code to save device number in OCSD. * Added code to read cold boot register signature to determine when BIOS should update/restore virtual Mac address. * Fixed bug where retry count was not always being updated. * Added code to support ECHO lopback command. * Added support NCSI Flex Address.
Version 3.20.00 Date: 8/30/2013 BIOS 3.0 version 3.22 uEFI version 5.27 FC Firmware - v7.00.01
- Added code to display message when 64 bit address detected. - Added support for Execution Throttle for P3P (ER106372) - Added code to not display connection options for Hilda Trinity adapter. - Check for invalid LUN in CTRL-Q scan devices. - Fixed bug in CTRL_Q utility that could cause some cosmetic display issues. - Add Hilda CLP _hss support. - Configure ISP24xx if on slot 0. - Added "QLogic" to PnP productid (ER104424) - Check for 4G flag in NVRAM for Hilda FC.(ER104714) - Fixed bug with CTRL-Q Loopback Utility (ER100729). - Added code to check/program Feature Config checksum. - Fix bug that could cause buffer overrun in CTRL-Q with Report Lun command (ER100294).
Version: 2.01.26 Date: Initial Release BIOS 3.0 version 3.16 uEFI version 5.26 FC Firmware - v6.03.00 qlgc-lnvgy_fw_fc_26xx-4_72_12_windows_32-64_chg
=============================================================================== Emulex 16GB FC Firmware Package ===============================================================================
Firmware Version: 10.6.228.14 Supported On: Lenovo System x Rack and Flex
Problems Fixed: - None
Incremental Interoperability: - This software/firmware is only supported with Lenovo Branded System BIOS - xClarity firmware download on specific platforms (requires minimum IMM version-please consult Lenovo on the co-requisite)
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: 10.4.303.3 Supported On: System x Rack
Problems Fixed: - VF speed is showing wrongly in ethtool and Ocm. - Firmware download using ethtool -f command is failing with opcode 172-1 failed:status 100-208 - Can not set the DelayDeviceDiscovery to Max (255) by ASU tool - Ports intermittently logging in to Brocade switch as G-Port
Incremental Interoperability: - xHMC event notifications on specific platforms (requires minimum IMM version-please consult Lenovo on the co-requisite) - Lenovo branded versions of formerly IBM branded products
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)
Problems Fixed: - Blue screen after firmware flash to with virtualized WWPN
Incremental Interoperability: - xHMC event notifications on specific platforms (requires minimum IMM version-please consult Lenovo on the co-requisite) - Lenovo branded versions of formerly IBM branded products
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)
Incremental Interoperability: - Grantley based ITE
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: 10.2.377.18 Supported On: System x
Problems Fixed: - Link does not recover during IO + Port Cycling with FW 1.1.43.3 - BIOS versions return multiple values - LPe16k point to point (direct attached) connection not bootable in legacy BIOS - LPe16k point to point (direct attached) connection performance issues in legacy BIOS
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: 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 - Reported firwmare version in CMM is intermittently incorrect - Unable to set pause Flow control when connected to DCBX switch with PFC off - during cable pull testing a 40 second delay in link up experienced with Brocade switch ports - I/O completion and abort completion WQE are generated out of order - Unexpected WCQE in RQ CQ seen in RXRI mode while performing cable pulls
Incremental Interoperability: - Inventory support for the Flex xHMC (Chassis Hardware Management Console) - Crosslink & Express Lane on LPe16k based solutions
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
(C) Copyright International Business Machines Corporation 2008 - 2014. All rights reserved. US Government Users Restricted Rights - Use, duplication, or disclosure restricted by GSA ADP Schedule Contract with IBM Corporation.
intc-lnvgy_dd_nic_20_0-361839b_windows_32-64_chg
ServeRAID M Series SAS/SATA Controller Driver for Windows Server 2012R2/2012/2008R2/2008
Version 6.708.09.00 - Moderate Impact/Low Probability of Occurrence --------------------- Fixes: - Incorporated MR6.8 driver stream - Fixed an issue where the storport driver could touch Dma32BitAddresses when it should not (SCGCQ00789403) - Fixed an issue where a "Driver detected possible FW hang, halting FW" error may occur during S3 sleep/resume (SCGCQ00840279) - Fixed an issue where newly-converted JBOD drives may not be presented to the OS (SCGCQ00824116) - Fixed an issue where Windows may enter a bootloop after upgrading driver with "Host driver needs to be upgraded to enable extended LD support" in the controller event log (SCGCQ00819305) - Fixed an issue where controller may take a long time to perform Online Controller Reset (SCGCQ00824414, SCGCQ00850132) - Fixed an issue where hibernation may result in BSOD with bugcheck A0 in Windows 2012 R2 (SCGCQ00827000) - Fixed an issue where an unexpected shutdown/reboot may occur during attempt of S4 sleep (SCGCQ00837484) - Fixed an issue where hibernation/dump cases may fail (SCGCQ00790690) - Fixed an issue where an error referencing "line 1276 in file ../../raid/1078dma.c" may occur during Online Controller Reset with JBODs present (SCGCQ00829169) - Fixed an issue where IO may be coalesced even if coalescing is disabled (SCGCQ00838379) - Fixed an issue where the system may BSOD/reboot when a drive is made offline via MSM (SCGCQ00889336) - Fixed an issue where the system may hang or BSOD when performing a firmware update with Online Controller Reset afterwards (SCGCQ00881576) - Fixed an issue where a dirty shutdown could result in false Automatic Event Notifications occurring on next boot (SCGCQ00795404)
Enhancements - Added support for new Windows event logging within the driver (SCGCQ00498613) - Enabled support windows driver sending 1MB IO transfers to firmware (SCGCQ00816315) - Added a feature where an MFI command timeout is handled with an Online Controller Reset (SCGCQ00806523) - Updated queue depth calculation so the queue depth is split equally among LDs and PDs (SCGCQ00818399) - Added support for logging Critical events in Windows Event Logger (SCGCQ00878364)
Version 6.706.06.00 - Moderate Impact/Low Probability of Occurrence --------------------- Fixes: - Incorporated MegaRAID 6.6 Windows driver stream - Fixed an issue where write performance dropped in Write-Through mode, and read performance dropped in some situations (SCGCQ00751029) - Fixed an issue where Windows might BSOD on reboot after updating to out of box driver version 6.704.12.00 (SCGCQ00793073)
Enhancements: - Added MFI_STAT_CONFIG_SEQ_MISMATCH handling in drivers (SCGCQ00651432) - Added ability to read PCI Domain/Segment information via StoreLib (SCGCQ00712490) - Added ability for driver to differentiate between FastPath IO and LD IO in the PendingIOQueue (SCGCQ00726747) - Added support for 1MB IO if the firmware supports it as well (SCGCQ00712518) - Added support for Self-Encrypting Drives in JBOD mode (SCGCQ00711558) - Added support for ServeRAID M5225, Lenovo x240 M5 M1210e, Lenovo x880 M5 M1210e, Lenovo x3650 M4 M5110e, Lenovo x3750 M4 M5110e
Version 6.704.12.00 - Moderate Impact/Low Probability of Occurrence --------------------- Enhancement(s): - Implement PnP ID for 2208 ROMB
Version 6.702.07.00 - Initial release ------------
lnvgy_dd_sraidmr_6_708_09_00_windows_32-64_chg
Qlogci iSCSI Adapter STOR Miniport Driver for Windows Release Notes Package version 2.1.6.10 April 29, 2015
This software license applies only to QLogic customers. QLogic Corporation. All rights reserved.
Table of Contents
1. Version 2. Changes/Fixes/Enhancements 3. Known Issues 4. History 5. Additional Information 6. Embedded Firmware Details 7. Trademarks 8. Notices 9. Contacting Support
1. Version
These release notes describe the changes, fixes, known issues, and release details that apply to iSCSI driver, version 2.1.6.10.
NOTE: Before installing this driver, you must update the Microsoft STOR miniport driver.
The FC is supported with Microsoft Windows Server 2008, Windows 2008 R2, Windows 2012 and Windows 2012 R2.
2. Changes / Fixes / Enhancements
2.1.6.10 QL - 04/24/14 * version number update 2.1.6.10A26 QL - 02/26/14 * Changed the code to follow helga idc spec 0_2, the lowest function number driver instance takes precedence when multiple functions of the same protocol are active during reset-recovery. Scope: 8200 Series adapters. Resolution: Will check if is the lowest function before to own the reset-recovery. 2.1.6.10A25 QL - 02/21/14 * ER114595: iSCSI Minidump extracted from QConvergeConsole CLI generates 100mb+ file. Scope: 8200 Series adapters. Resolution: Will return the actual size of minidump captured memory to application. * Changed the code to follow helga minidump spec v0.94, the driver should keep the first firmware dump collected. Scope: 8200 Series adapters. Resolution: Will not overwrite the holding md buffer, until application retrieved it. 2.1.6.10A23 QL - 01/16/14 * ER114878: Firmware Dump state should be honored by the iSCSI driver once the state has been changed successfully Scope: 8200 Series adapters. Resolution: Driver is changed to support the minidump state. * ER114598: Adapter reset while enabling/disabling VLAN and Large Frames Scope: 8200 Series adapters. Resolution: Changed the driver to avoid sending down a session close for a already disconnected session. 2.1.6.10A22 QL - 11/14/13 * ER113419: Adapter reset when iSCSI disable enable is done after peg halt loop Scope: 8200 Series adapters. Resolution: Fixed a bug in Interrupt enable. 2.1.6.10A20 QL - 11/13/13 * Change the code for GET_MANAGEMENT_DATA to follow the spec Scope: 8200 Series adapters. 2.1.6.10A19 QL - 10/23/13 * Changed the sequence of tasks for stop adapter Scope: 8200 Series adapters. 2.1.6.10A18 QL - 10/17/13 * ER112269: "The driver detected a controller error on \Device\RaidPort3" message seen in event log on add/remove isns. Scope: 8200 Series adapters. Resolution: Fixed a bug in Link AEN processing. 2.1.6.10A17 QL - 10/11/13 * ER112232: Helga adapter has been reset when iSCSI Driver is loaded. Scope: 8200 Series adapters. Resolution: Fixed a bug in Quiscent supporting. 2.1.6.10A16 QL - 9/19/13 * ER111214: System BSoD when FCoE external loopback is run with loopback cable connected Scope: 8200 Series adapters. Resolution: Driver is changed to limit the number of link status checking mailbox command. * ER111448: Initiator port does not list the unbinded targets after logging in to it's iSNS targets Scope: 8200 Series adapters. Resolution: Fixed driver bug, which could cause a iSNS protocol error when it receive a SCN frame while doing targets scanning. 2.1.6.10A15 QL - 8/28/13 * Mini-dump creation ioctl added Scope: 8200 Series adapters. Resolution: Driver is changed to support the ioctl INT_CC_HBA_CMD_MINIDUMP. * Implemented fast devices scanning logic in initialization Scope: 8200 Series adapters. Resolution: Driver is changed to use the device index information from the FW to scan for the next configured device instead of blindly scanning all devices. This will improve the device scanning time during initialization. 2.1.6.10A13 QL - 8/05/13 * ER109767: Continuous adapter reset while assigning static IP to the iSCSI adapter ports Scope: 8200 Series adapters. Resolution: Changed the driver to avoid sending down a session close for a already disconnected session. 2.1.6.10A11 QL - 7/11/13 * fixed a bug for whql crash dump Scope: 8200 Series adapters. Resolution: Driver was using too much uncached memory as a crash dump driver, changed the driver to avoid allocate unnecessary uncached memory in crash dump mode. 2.1.6.10A10 QL - 7/08/13 * send down stopfw for reset Scope: 8200 Series adapters. Resolution: Changed the driver to send down stopfw for reset, when detected fw alive count updating stopped, but not yet continued 3 or 5 seconds. 2.1.6.10A6 QL - 5/29/13 * ioctl EXT_CC_GET_AEN could go through even adapter is not in ready state Scope: 8200 Series adapters. Resolution: Changed the driver to allow the EXT_CC_GET_AEN go through even adapter is not in ready state. * added Full 64 bit addressing support Scope: 8200 Series adapters. Resolution: Changed the driver to support full 64bits support. * deleted the system log for need reset detected Scope: 8200 Series adapters. Resolution: Changed the driver to avoid the unnecessary log event. * set md mask default to be 0x1 Scope: 8200 Series adapters. Resolution: Changed the driver to set the minidump mask to be 0x1 as default. * added a force md ioctl Scope: 8200 Series adapters. Resolution: Added a ioctl to allow an application to force a minidump. 2.1.6.10A5 QL - 5/16/13 * set the md to be disabled as default Scope: 8200 Series adapters. Resolution: Changed the driver to set the minidump to be disabled as default. 2.1.6.10A4 QL - 4/24/13 * fixed a scsi pass through ioctl bug Scope: 8200 Series adapters. Resolution: Driver is treating the SCSI pass through IOCTL as a normal IO command thus causing the system to panic when accessing the buffer address at a higher priority. The change is to check the IOCTL command flag correctly and handle it via the IOCTL path. 2.1.6.10A3 QL - 4/22/13 * Implemented fast devices scanning logic Scope: 8200 Series adapters. Resolution: Driver is changed to use the device index information from the FW to scan for the next configured device instead of blindly scanning all devices. This will improve the device scanning time during recovery. * Added a registry key "mrdt" to control the max response queue dpc processing time Scope: 8200 Series adapters. Resolution: Changed the driver to limit the respones queue processing time in the DPC routine. * Added a registry key "qd" to control the queue depth Scope: 8200 Series adapters. Resolution: Added queue depth address configuration. 2.1.6.10A2 QL - 3/25/13 * Added ip registry key Scope: 8200 Series adapters. Resolution: Added IP address configuration.
3. Known Issues
N/A
4. History
N/A
5. Additional Information
None
6. Embedded Firmware Details
This section provides information for the firmware embedded in the driver:
* Not available
7. Trademarks
QLogic, the QLogic logo, QLA, SANbox, SANsurfer, and SANblade are registered trademarks of QLogic Corporation.
SANtrack and SAN Express are trademarks of QLogic Corporation.
All other brand and product names are trademarks or registered trademarks of their respective owners.
8. 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.
9. Contacting Support
Please feel free to contact your QLogic approved reseller or QLogic Technical Support at any phase of integration for assistance. QLogic Technical Support can be reached by the following methods:
North America Contact Information
Email: support@qlogic.com
Phone: (952) 932-4040
Web: http://support.qlogic.com
Support contact information for other regions of the world is available at the QLogic website: http://support.qlogic.com
(c) Copyright 2015. All rights reserved worldwide. QLogic, the QLogic logo, and the Powered by QLogic logo are registered trademarks of QLogic Corporation. All other brand and product names are trademarks or registered trademarks of their respective owners. qlgc-lnvgy_dd_iscsi_2_1_6_10b_windows_32-64_chg
=============================================================================== Emulex FC Device Driver for Windows ===============================================================================
Device Driver Version: 10.6.114.0 Supported On: Lenovo System x Rack and Flex
Problems Fixed: - Port Beaconing feature not working on Windows with Brocade 16G Switch
Incremental Interoperability - Add Queue Dump in driver memory region to help debugging
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 - XE102/XE104 - OCManager reports link down on both ports
Incremental Interoperability: - Crosslink & Express Lane on LPe16k based solutions
Device Driver Version: 10.0.720.0 Supported On: IBM System x
Problems Fixed: - Concurrent Hardware and Operating System (CHAOS) Test fails in XE-10x 10G - Extloopback command not supported with LPe16000 adapters with 16Gb link speed - Windows FC driver bug with Data Domain virtual tape library
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
UEFI Flash Update Version 1.20 - BuildID TAE108F Change History
Version 1.20 - BuildID: TAE108F ------------------------------------------------------------------------------ Problem(s) Fixed: 1. Fix SBE can't be trigger issue. 2. Fix Downgrade Issue from Broadwell to Haswell. 3. Fix the issue that reboot command can't be used after boot to Emulex CD-ROM. 4. Fix The behavior of Deployment mode is abnorma1 under OneCLI. 5. Fix TPM POST message will not display in (rare) condition. 6. Fix VSS POST message will not display in (rare) condition. 7. Fix P-state abnormal behavior under OneCLI. 8. Fix wrong behavior of third Nx POST function. Enhancements: 1. Update Intel RC to 1.17 2. Update microcode to m6f306f2_00000035.inc 3. Update Intel ACM to 2.2 4. Improve boot performance Limitations: Dependencies:
Version 1.12 - BuildID: TAE106K ------------------------------------------------------------------------------ Problem(s) Fixed: 1. Fixed PCIe AER error on solarflare card installed on slot 2 Enhancements: Limitations: Dependencies:
Version 1.11 - BuildID: TAE106J ------------------------------------------------------------------------------ Problem(s) Fixed: 1. Fixed a Processor Trusted Execution Technology(TXT) cannot be enabled issue 2. Fixed that diag Code [S.1100B] detailed content is empty in uEFI SEL 3. Fixed that ASU tool Cannot change OperatingModes.TurboMode 4. Fixed that onboard devices setting duplicated in network device list 5. Fixed that excessive memory correctable errors Logging Limit (PFA Threshold) Enhancements: 1. Support Memory Reference Code Release version 1.13 2. Enhance security to prevent the installation of non-genuine code onto the system. Limitations: Dependencies:
=============================================== Change History =============================================== Document Name: Change History Document Status: Lenovo General Avalibility Document Version: 3.2.5 Latest Driver Version: 3.2.5.0 Document Updated: 09/22/2014 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 "Lenovo 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 Lenovo 16G Mezzanine Fabric Adapter.
HBA - Fibre Channel 16G Support
Fixes and Enhancements: N/A brcd-lnvgy_dd_cna_3_2_5_0_windows_32-64_chg
=============================================================================== Emulex FCoE Device Driver for Windows ===============================================================================
Device Driver Version: 10.6.114.0 Supported On: Lenovo System x Rack and Flex
Problems Fixed: - Port Beaconing feature not working on Windows with Brocade 16G Switch
Incremental Interoperability - Add Queue Dump in driver memory region to help debugging - Flex System CN4052 R 2-port 10Gb Virtual Fabric Adapter - Lenovo 00AG572 VFA5 2x10GbE SFP+ PCIe Adapter - Lenovo 00AG582 VFA5 2x10GbE SFP+ PCIe FCoE/iSCSI Adapter - Lenovo 00AG562 VFA5 ML2 2x10GbE SFP+ Adapter
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 - XE102/XE104 - OCManager reports link down on both ports
Device Driver Version: 10.0.720.0 Supported On: IBM System x
Problems Fixed: - Concurrent Hardware and Operating System (CHAOS) Test fails in XE-10x 10G - Windows FC driver bug with Data Domain virtual tape library
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
Broadcom (R) Corporation Broadcom Ethernet Software Kit RELEASE 17.0.0.4 from updated Gold CD 17.0.4.4 (March 31, 2015)
RELEASE NOTES
Summary ======= - Gold Release for the Broadcom 17.0 project.
Caveats ======= - Limited testing on some OSes
Highlights ========== - Updated 64-bit parameter to get target CPU for DPC due to link not recognized on system with 64 cores. - Updated the driver id in APE host memory when system resuming from S4. - Add code change to make driver handle interrupt message affinity correctly - Fixed issue where pluggin in two 5722 NICs on the same system, cause one of the NICs to sometimes not function. Fixed issue where device shows link down randomly and is unable to pass traffic. - Fixed unexpected rebooted on PCI Root Port Surprise Remove Test (PCI device only) - Fixed BSOD while running BACS Diagnostics stress on 5719 - Added change to force an interrupt if rx traffic is idel and interrupt is unexpectedly disabled. - Enhanced performance by changing driver to indicate multiple packets and insert multiple RBDs to device at a time.
Operating System coverage: ========================== Windows 2008 R2 Windows 2012 Windows 2012 R2
Version history between updated CDs v17.0.4.4 release and previous CD v16.8.4.1 release: ====================================================++++++================
CD Release Updated v17.0.4.4 Previous v16.8.4.1 Product Version Version
Windows Server 2008/2012/2012R2 x64 Driver 17.0.0.4* 16.8.0.3 Windows Server 2008/2012/2012R2 IA32 Driver 17.0.0.4* 16.8.0.3
Note: An asterisk (*) following a version number represents a certified driver.
Windows 2000/2003 is not Supported with this package.
This release contains the following software programs:
NetXtreme Gigabit Ethernet NDIS6.x for Microsoft Windows 2008/2012/2012R2, 32 bit extensions - b57nd60x.sys NetXtreme Gigabit Ethernet NDIS6.x for Microsoft Windows 2008/2012/2012R2, 64 bit extensions - b57nd60a.sys
* Third-party trademarks and copyrights are the property of their respective owners.
Broadcom, Broadcom Advanced Server Program, Broadcom Advanced Server Control Suite, and Smart Load Balance are registered trademarks of Broadcom Corporation.
brcm-lnvgy_dd_nic_17_0_0_4_windows_32-64_chg
QLogic 10 GbE Converged Network Adapter MultiFlash Update for System x Lenovo P3P Firmware Change History QLogic Corporation. All rights reserved.
Package Version 2.80.02.a
Updated QCCCLI
Release Version: 2.80.02
Table of Contents 1. Version 2. Changes 3. Limitations 4. Known Issues 5. Additional Information 6. Trademarks 7. Notices 8. Contacting Support
1. Version
The files contained in this Flash image package are zipped into a file that will expand to provide the following versions for the 3200/8200 Series Adapters.
* Flash Image Version 2.80.02
P3P28002.BIN contains: ---------------------- * NIC PXE (3200/8200 Series) PXE 3.0 version 2.0.2.37 PXE EFi version 2.6.21 Signed * FCoE BIOS (8200 Series only) BIOS 3.0 version 3.29 uEFI version 6.09 Signed * iSCSI BIOS (8200 Series only) BIOS 3.0 version 2.21 uEFI version 2.04.12 signed * Firmware - v4.16.153 * Pegtune - v4.16.32 * Bootloader - v4.16.05 * CRBinit - v4.12.07
Updated QCCCCLI to support windows WS2K8 R2; Package Version 2.20.18.b.
Hardware Support Standard Servers with PCIe 1.1 and 2.0 x8
3. Limitations
None
Industry Standards Compliance N/A
4. Known Issues
5. Additional Information None
6. 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.
7. 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.
8. Contacting Support Please feel free to contact your QLogic Technical Support at any phase of integration for assistance. QLogic Technical Support can be reached at:
Gary Stift gary.stift@qlogic.com
(c) Copyright 2011. All rights reserved worldwide. QLogic, the QLogic logo, and the Powered by QLogic logo are registered trademarks of QLogic Corporation. All other brand and product names are trademarks or registered trademarks of their respective owners.
=============================================================================== Mellanox OFED for Windows UpdateXpress Release Notes for Lenovo Version MLNX_WinOF 4.95 Last Modified on Jun 02, 2015 ===============================================================================
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 v4.90 Main Changes (from v4.80) 4.2 WinOF VPI v4.80 Main Changes (from v4.70) 4.3 WinOF VPI v4.70 Main Changes (from v4.61) 4.4 WinOF VPI v4.61 Main Changes (from v4.60) 4.5 WinOF VPI v4.60 Main Changes (from v4.55) 4.6 WinOF VPI v4.55 Main Changes (from v4.40) 4.7 WinOF VPI v4.40 Main Changes (from v4.3) 4.8 WinOF VPI v4.3 Main Changes (from v4.2) 4.9 WinOF VPI v4.2 Main Changes 5. Known Issues 6. Bug Fixes History 6.1 WinOF VPI v4.90 Bug Fixes 6.2 WinOF VPI v4.80 Bug Fixes 6.3 WinOF VPI v4.70 Bug Fixes 6.4 WinOF VPI v4.61 Bug Fixes 6.5 WinOF VPI v4.60 Bug Fixes 6.6 WinOF VPI v4.55 Bug Fixes 6.7 WinOF VPI v4.40 Bug Fixes 6.8 WinOF VPI v4.2 Bug Fixes 7. API Changes 7.1 WinOF VPI v4.95 API Changes: 7.2 WinOF VPI v4.90 API Changes: 7.3 WinOF VPI v4.80 API Changes: 7.4 WinOF VPI v4.70 API Changes: 7.5 WinOF VPI v4.60 API Changes: 7.6 WinOF VPI v4.55 API Changes: 7.7 WinOF VPI v4.40 API Changes:
=============================================================================== 1. Overview =============================================================================== This document addresses the Mellanox WinOF VPI driver package distributed for Windows Server 2008 R2 (x64 only) and 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 10/40/56GB Ethernet ports. The port type is determined upon boot based on card’s capability and user setting.
=============================================================================== 2. Changes in This Release =============================================================================== InfiniBand -Fixed BSOD on next driver restart when running the sminfo tool and SM is not running -Fixed instability in IPoIB driver when polling mode is enabled -Fixed the issue when live-migrated VM with virtual NIC over IPoIB physical interface loses its IP address and comes up with APIPA address (169.254.x.x) Ethernet -Modified counter names in perfmon to be more clear: -"Sent Pause Frames" --> "Pause Frames Sent" -"Sent Pause Duration" --> "Total Transmitter Pause Duration (us)" -"Rcv Pause Frames" --> "Pause Frames Received" -"Rcv Pause Duration" --> "Total Requested Pause Duration (us)"
-Fixed BSoD after the receive buffer’s size changes in VMQ mode -ECN is now configurable via PowerShell -Fixed Powershell setting of RoCE mode when the machine has more than one Mellanox device -Fixed reporting of NVGRE capabilities to the OS -Added a new mode that ignores FCS warnings and enables the Ethernet packets to be received by the NIC -Added the option of setting the MAC address of 2008R2 teaming driver to be taken from the primary interface -Added the option of taking the MAC Address used for teaming from one of the interfaces without manipulation UI -Added the option of configuring the team MAC address to be the same as the primary adapter MAC address Troubleshooting -Added an autologger session that dumps WPP traces to file to enable -Added logging of performance counters and CPU power saving state to system snapshot tool RDMA -Fixed handling of send request with inline data larger than supported -Added the option to allow RDMA programmers to create pre-allocated pools of ND resources to reduce resource creation time Low Level Driver -Fixed the case in which the system rarely gets unstable after driver restart Infrastructure -Updated customization for OEM cards
=============================================================================== 3. Supported Firmware and Firmware Tools =============================================================================== - Firmware: - ConnectX®-3 and ConnectX®-3 EN QDR/FDR10/FDR - Firmware v2.34.5000 and above
- Tools: - MFT (Mellanox Firmware Tools) version 3.0.0 and higher
=============================================================================== 4. Change Log History ===============================================================================
4.1 WinOF VPI v4.90 Main Changes (from v4.80): ------------------------------------------------------------------------------- WinOF VPI v4.90 provides the following changes and new features:
Generic -Added support for new Mellanox OUI: E4:1D:2D. Mellanox card with this OUI would not work with the previous versions of WinOF -Added to MLNX_System_Snapshot Mellanox specific counters and data from Get- Mlnx* Cmdlets Resiliency -Reset Flow improvements: redesigned and simplified it to resolve race conditions with parallel resets Ethernet -Added ETS with DSCP support -Improved completion memory access speed -Improved RX descriptors memory access -Improved the scalability of VMQ to allow running more VMs reliably -Enabled ETS per flow according to the DSCP value in the IP header -Reduced memory footprints of Ethernet interfaces -Changed default VMQ/VPort affinity to use the first RSS CPU -Changed behavior in VMQ/SR-IOV mode by making all traffic of the default VPORT/ VMQ use the same CPU InfiniBand -Updated IBAL interface version. In order for the applications that use the IBAL interface to work with WinOF Rev 4.90.50000, they must be recompiled with the new SDK -Added support for SM change event RoCE -In RoCE v2, added the option of determining the source port field of the UDP header can by the application NDK -Improved CPU utilization -Changed ndkgetremotetokenfrommr() to return value in network byte order IPoIB -Added support for multiple PKey interfaces in IPoIB -Added support for teaming of IPoIB interfaces to allow failover -Added IPoIB adapters teaming support (beta level) -Added sending of gratuitous ARP in IPoIB interface when the MAC address is changed -Reduced memory footprints of IPoIB interfaces -Reduced the multiple number of path record queries to one when old query information exists -Improved completion memory access speed -Changed default VMQ/VPort affinity to use first RSS CPU -Multiple PKey support is now at GA level. The part_man tool allows the creation of up to 64 vIPoIB interfaces (32 per port) -Added a warning to the event log if the port MTU is higher than the reported MTU by the SM. ND -Increased the number of supported SGEs in 32 bit DLLs to 2 Changes in UI -Replaced the terms “LBFO” and “Bundle” with “Teaming” and “team” respectively.
4.2 WinOF VPI v4.80 Main Changes (from v4.70): ------------------------------------------------------------------------------- WinOF VPI v4.80 provides the following changes and new features:
- Generic - Changed Reset Flow (+SR-IOV)-enabled only if no user space application is running and depends on the registry key: AllowResetOnError setting) - Changed the number of supported QPs in a multicast group from hard coded value to firmware capabilities dependent - Optimized handling of “affinity change” on OID_RECEIVE_FILTER_QUEUE_PARAMETERS - Added the ability to control the number of retries and timeout to check the device health before performing reset.
- IPoIB - Added multiple P_Key support (beta level) - Added IPoIB SR-IOV over KVM and ESX Hypervisors (for both full and partial mem- bership) - Added support for LID change event - Added enhancements in part_man for the multiple Pkey support - Changed IPv6 “all dhcp servers” mcast to be persistent
- Ethernet - Added RSS in UDP (enabled by default) - Added 56 GbE (Please refer to the Infiniband Switch User Guide for further details) - Changed DSCP configuration to be per port instead of global - Network Direct: Fixed race in NDK between handling of incoming connection and destruction of a listener - Network Direct: Fixed race between NDK object creation and usage - Improved TCB (Transmission Control Block) management on send - Improved transmit and receive in multi stream scenarios - Enabled hardware checksum offload for non TCP/UDP traffic with ConnectX3-Pro - Improved stability when handling OIDs during driver reset
- RoCE - Added IP based RoCE (v1.25)
- Infiniband - Added non-default PKey in VM
- Performance - Optimized interrupt moderation values in SR-IOV VF mode for IPoIB - Improved perf_tuning detection for the first port - Improved performance in packet forwarding scenarios - Decreased dropped packets rate for Ethernet significantly - Changed default perf_tuning scenario to be “Balanced configuration” - Various performance improvements
- WMI/CIM - Added ability to read active RoCE configuration from hardware - Added support for RoCE v1.25
- Installation/ Upgrade - Added check for administrator privileges during installation - Added support for installation in silent mode without execution of perf_tune - Fixed installation stuck when Remote Desktop Session Host Windows Installer RDS compatibility is enabled
4.3 WinOF VPI v4.70 Main Changes (from v4.61): ------------------------------------------------------------------------------- - Generic - Changed Ethernet and IPoIB event log messages to be more clear - Ported SDK project to Visual Studio 2013.
- Performance - Improved perf_tuning setting in single CPU machines to avoid TX collision
- IPoIB - Added support for IPoIB SR-IOV Virtual Function (VF) over KVM Hypervisor (Beta level) - Added support for non-default pkey, as queried from OpenSM, on IPoIB SR-IOV VF over KVM. - Added IPoIB QoS proprietary counters, diagnostics and traffic for monitoring, using Windows' perfmon utility
- Ethernet - Added Virtual Ethernet Adapter support which enables using SMB Direct and HyperV (VMQ and NVGRE (over ConnectX®-3 Pro)) on the same port - Added lossless TCP buffer management when no receive WQE are available
- RoCE - Added ConnectX®-3 Pro support for RoCEv2 - Changed the transport name in vstat and ibstat to be RoCE v2.0
- Installation/Upgrade - All user-space binaries are now signed - Fixed restoration process of DNS servers during upgrade - Fixed popping windows during installation/upgrade - Fixed missing 32 bit files in the catalog files
4.4 WinOF VPI v4.61 Main Changes (from v4.60): ------------------------------------------------------------------------------- - 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
4.5 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
4.6 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
4.7 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
4.8 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
4.9 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 ===============================================================================
- Generic - 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 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. - 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. - The UI displays the network interface’s state incorrectly even after disabling and re-enabling the "disconnected" network interfaces. To see the correct state and link speed, perform one of the following: - Run Get-netadapter Powershell cmdlet - Right click on that interface from “Network Con- nections” and click on status - WoL is not supported on Windows 2008 R2. - iSCSI boot over Windows 2008 R2 is not supported - Clearing the Mellanox device counters through perfmon does not always work.To resolve it user should restart the driver - When running in SRIOV mode and Hyper-V, the same driver version should be used on both the Hypervisor and the virtual machine. - Working with user space RDMA resources (i.e. ND) in parallel with hardware reset may cause unexpected behavior. - In Windows Server 2008 R2 and Windows 7 Client, ConnectX-3 Pro is identified as ConnectX-3. This does not affect anything. - 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. - The drivers’ functionality is limited up to 128 cores. - IPv6 traffic between Hyper-V hosts over IPoIB v-Switch may experience traffic loss. - Creating a virtual IPoIB port with non-default PKey isn't supported in the WinOF upgrade. Doing so will indicate that the adapter is unplugged after the upgrade ends. - InfiniBand application that was compiled with an SDK version earlier than WinOF v4.90 is not binary compatible. - 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”. - 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. - 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. - IPoIB does not support: - MAC address change - QoS (packet priority) - Connected Mode - 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. - 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. - 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. - Burning a Firmware image with a "mtu_cap" value other than the default causes the driver load to fail. - Calling ib_query_cq() on a CQ which was created with 1 entry indicates that there are 0 entries in the CQ. - IPoIB ports report the configured MTU instead of the effective MTU. These MTUs differ when the corresponding InfiniBand partition is configured with a smaller MTU than the port’s configured MTU. In such case, there may be communication failures and/or CPU work overhead on the communicating nodes (even if the peer node is configured correctly). - The displayed MAC address in the DHCP server of virtual IPoIB I/F may display a wrong data (FF00........) although the I/F is still fully functional. - DHCP messages that IPoIB send are not fully spec complaint. The spec specifies that the 'chaddr' field must be zeroed, but WinOF IPoIB does not guarantee it.
- Ethernet - 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". - 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. - 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 - 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. - Disabling the ethernet adapter with more than 25 VLANs configured over windows server 2008 R2 \ windows server 7, may result with a non-responding server. - Virtual Ethernet Interfaces created by vea_man are not tuned by the automatic performance tuning script - In Windows Server 2008 R2, devices created by the Teaming driver do not show the correct OEM branding in the Device Manager. - Wake on Lan (WoL) cannot be disabled on NICs which supports it.
- Quality of Service - 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 NVGRE - RoCE does not support: - Traffic cannot go through the router. It works in the same subnet only - Multicast traffic - VLAN - Layer 3 feature - In machines with heterogeneous NICs: a NIC which sup- ports RoCE v2, and a NIC which does not support RoCE v2 the following issus might raise: - Connect-X®-3 Pro is loaded with the transport type RoCE v2 - Connect-X®-3 is loaded with the transport type RoCE - 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. - Using different versions of RoCE in your cluster is not supported. - RDMA Activity counters do not count during NetworkDirect RoCE traffic. - GRE traffic steering by inner MAC and by outer MAC simultaneously is currently not supported. - 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.
- Performance - 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. - 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. - The driver uses optimal interrupt moderation values for 10 GbE SR-IOV VF scenario. For other scenarios, the optimal values yet to be found. - While running in a Virtual Machine (working with VF) or in a Native Machine, performance counters, if read directly or by using any tool (as Windows Task Manager), may show that no packet has been sent/received. This happens because the driver periodically examines the actual performance counters and caches the results. If the sample rate is too high, the counter values will remain the same.
- Hyper-V - 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. - 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" - SR-IOV - Working with Windows guest OS over non-Windows SR-IOV hypervisor may result with higher latency compared to Windows Hypervisor. - When working with OEM adapter cards in SR-IOV mode, the VF is identified as a Mellanox adapter. There is no other effect on the behavior of the VF. user should verify that the FW INI contains the parameter vf_subsystem_id with the OEM encoding in the [HCA] section. - An SR-IOV Virtual Machine is loaded in InfiniBand mode when no OpenSM is available in the subnet. The following event appears in the event log: “detected a null port GUID for port . A Virtual Function device may have a null port GUID if there is no OpenSM instance on its network. Please make sure the network has an active OpenSM and restart the driver.” User should start the OpenSM and restart the driver. - For InfiniBand SR-IOV guest, OpenSM Assigned GUIDs are not supported and may cause unexpected behavior User should work only with Administrator assigned GUIDs. - For InfiniBand SR-IOV guest, ND traffic does not work on an SR-IOV adapter when a Para-Virtualization adapter configured on the same virtual machine and the same subnet as IPoIB. User should disable the Para-Virtualization adapter. - In SR-IOV mode over Hyper-V, all ports are in pure Ethernet mode and RDMA is not supported on either port. - In SR-IOV mode, enabling SR-IOV in the BIOS may change the interfaces names. If any VSwitch is bounded to an interface whose name was changed, there will not be any way to perform any operation on them. User should unbond all Vswitches from the NIC's interfaces before enabling SR-IOV in the BIOS. - In SR-IOV, 40Gbps interfaces of vSwitch interface on HyperV, and the VMNIC are identified as 10Gbps. Despite the incorrect speed identification, they still achieve 40Gbps performance. - Device name in the Device Manager does not show the correct OEM branding for SR-IOV Virtual Function devices without the updated firmware - Installation / Upgrade - 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. - 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 - 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. - Running a downgrade in silent mode is not supported. Upon downgrade the return code will always be 0. - Uninstalling the driver after upgrade won't remove the directory %ProgramFiles%\Mellanox - Uninstalling the driver when multiple of VLANs are con- figured never ends. - mlxfwmanager which is supplied in the package is not signed. - 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. - Installation/upgrade fails due to PNP failure to copy the driver files to the driver store, and the following text is printed in the event logs: Fault bucket, type 0 Event Name: PnPDriverImportError Response: Not available Attached files: C:\Users\\App- Data\Local\Temp\DMI151A.tmp.log.xml C:\Program Files\Mella- nox\MLNX_VPI\ETH\mlx4eth63.inf To solve it reboot the machine and reinstall. - Installation/upgrade fails due to failure to stop the WMI service. The following text is printed in the installation log:“CustomAction StopWMIService returned actual error code 1603” - Users should kill the WMIPrvSE.exe tasks in the task manager and reinstall. - Following the upgrade of Mellanox driver to WinOF-4.60 and above or on servers with no internet access, the first PowerShell command might be stuck for ~2- 3 minutes before its completed. According to the following content, this issue is related to .Net framework version or an issue with the internet access:http://www.minasi.com/forum/ topic.asp?TOPIC_ID=39253. users should Run the following script on the server to optimizes loading PowerShell DLLs: $Env:PATH = [Runtime.InteropSer- vices.RuntimeEnvironment]::GetRuntime- Directory() [AppDomain]::CurrentDomain.GetAssem- blies() | % { $pt = $_.Location if (! $pt) {continue} if ($cn++) {''} $na = Split-Path -Leaf $pt Write-Host -ForegroundColor Yellow "NGENing $na" ngen install $pt} - On ConnectX-3 cards only, when upgrading from Windows 2012R2 Inbox driver to WinOF, the RoCE mode setting in the registry is not properly transferred to the new driver. In case a non-default value was used it will not be configured following the upgrade. - Uninstall from the Device Manager is currently not supported - 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. - 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. - If there are disabled network interfaces or a disabled Mellanox bus driver, they will be enabled after the WinOF upgrade. - The upgrade from WinOF v4.58 to WinOF v4.90 saved only the configuration of the Ethernet interface, and did not save the configuration of the IPoIB interface.
- Utilities - 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 over- head of creating extra copy for each packet. This may lead to packet drops on the link. users should verify Ethernet flow control is enable to ensure a lossless link. - Pcap file issues for RoCE v1.5: - The packets 'capture-time' field is wrong (may be 0 or negative numbers). - For every captured packet, an additional 0 size flame is added. This appears in Wireshark as a 'malformed Ethernet packet'. - Sniffing over IB ports is currently not supported. - Creating virtual adapters over IPoIB using part_man is currently not supported in all OEM adapter cards. - When the tracer tool traces many events, it may consume a large amount of memory (up to several GB RAM).
- CIM/WMI - Running Microsoft CIM cmdlets operations and their derived classes on classes MLNX_NetAdapterSettingData and MLNX_NetAdapterRoceSettingData is not sup- ported. Calling those commands may cause the debugger, if connected to the machine, to assert. Users should use DriverCoreSettings instead. - For PCI Gen3, PcieLinkSpeed is reported as “Unknown” when running Get-NetAdapterHard- wareInfo Powershell cmdlet. - WMI does not work due to lack of permissions.Users should change the execution policy by running the following command: Set-ExecutionPolicy AllSigned -The information that is printed in the cmdlets get-netadaptersriov and Get-MlnxPCIDeviceSetting is inconsistent. Users should use only the cmdlet Get-MlnxPCIDeviceSetting
- ND - There is an interoperability problem between NDK and ND in RDMA operations during the write and read operations. However, the send operation resumes working. This happens since ND uses remote token in a network order (big endian) while NDK uses remote token in CPU order (little endian). Therefore, an inconsistency is caused between ND and NDK in RDMA operations. An ND application that works with NDK using RDMA operations must handle this issue by changing the remote token to the appropriate byte order before sending it to NDK. - 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.
=============================================================================== 6. Bug Fixes History =============================================================================== 6.1 WinOF VPI v4.90 Bug Fixes: ------------------------------------------------------------------------------- -Fixed CM CPU consumption for a long period of time while in DPC context -Fixed stability issues -Fixed failure when deleting Virtual Ethernet Interfaces right after creating it -Fixed duplicated values of Receive Completion Method in UI on Windows Server 2012 R2 -Fixed propagation of error code when ib_join_mcast() fails -Fixed connectivity problems when using PKeys from the same partition with different membership types -Fixed VM reset after printing the message "mlx4_core 0000:05:00.0: unparavirt com- mand: OTHER (0x3a) accepted from slave:3" in SR-IOV InfiniBand VM over non-win- dows hypervisor -Fixed the UI crash when working with a single port -Fixed wrong ND_* performance tests duration calculation -Fixed the higher limit of ThreadPoll to be 200,000 -Fixed stability issues -Fixed displaying of IPoIB default turning option -Fixed temporary network connectivity issues while migrating VMs or modifying VMQ configuration for VMs that uses IPoIB with VMQ -Fixed the part_man tool to use the actual default p_key instead of 0xffff -Fixed NIC reset when attaching to a multicast group fails -Fixed duplicated values of Receive Completion Method in UI on Windows Server 2012 R2 -Fixed CIM failure after installation in maintenance mode -Fixed loading of old driver after driver upgrade that requires system reboot to complete the process -Fixed RoCE disable by default after installation of WinOF in Windows 8.1 Client -Fixed seg fault when executing ND application with no device installed or when a wrong device identifier is used -Fixed wrong reported value of supported number of SGE in 32 bit DLLs -Fixed restoration of NVGRE configuration after NIC reset -Fixed failure when calling Get-MlnxFirmwareIdentity if not all devices are up
6.2 WinOF VPI v4.80 Bug Fixes: ------------------------------------------------------------------------------- - Fixed driver load failure in machines with 1 TB memory and above - Fixed rare cases of driver hang following a Subnet Manager failover event - Fixed SM fail-over causing the driver to hang - Fixed stability issues - Fixed performance tuning for 1GbE link - Fixed possible reset of driver during migration of large number of VMs at the same time - Fixed stability issues - Fixed missing pause response by sender when using DSCP/untag priority tag mode with ETS enabled - Fixed wrong return value in IND2Adapter::QueryAddressList - Fixed installation stuck when Remote Desktop Session Host Windows Installer RDS compatibility is enabled
6.3 WinOF VPI v4.70 Bug Fixes: ------------------------------------------------------------------------------- - Fixed an issue which caused Mellanox miniport devices to be listed in “Devices and Printers” - Fixed Ethernet and IPoIB deadlock in power state change during shutdown/reboot - Fixed stability issues - Fixed part_man exit with return value 0 in case of error - Fixed ibstat behavior on devices with RoCE enabled - Fixed releasing of RDMA resources and reacquire them on power down and up. - Fixed RDMA Activity counters which didn't increase for ND traffic - Fixed hard-coded limitation of 4 SGEs - Fixed vstat printing of counters for Ethernet ports - Fixed removal of virtual IPoIB ports in uninstallation
6.4 WinOF VPI v4.61 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
6.5 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
6.6 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
6.7 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
6.8 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 v4.95 API Changes: -------------------------------------------------------------------------------
7.2 WinOF VPI v4.90 API Changes: ------------------------------------------------------------------------------- struct ib_mad_element_t - Added the field rroce_udp_s_port struct ib_av_attr_t - Added the field rroce_udp_s_port VERBS_MINOR_VER - Increased its value, 0x00c -> 000d
7.3 WinOF VPI v4.80 API Changes: ------------------------------------------------------------------------------- - RDMA_TRANSPORT_RDMAOE_1 is an alias to RDMA_TRANSPORT_RDMAOE - Added enumerated values RDMA_TRANSPORT_RDMAOE_1_25 - Added new functions is_roce(), is_mac_based_roce(), is_ip_based_roce(), is_rroce_or_ip_based_roce() - p_next in struct ib_wc_t was replaced with an anonymous union which contains two fields: p_next and qp_context
7.4 WinOF VPI v4.70 API Changes: ------------------------------------------------------------------------------- - Added a new function ib_get_port_spl_qp() - Changed API (one more input parameter was added to ib_get_mad_inner() - Changed API (one more input parameter was added to ib_get_mad() - VERBS_MINOR_VER value was changed from 000a to 000c - Added a new macro for UNBOUND_PORT_NUM
7.5 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
7.6 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
7.7 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-lnvgy_fw_nic_4_95_50000p1_windows_x86-64_chg
Dynamic System Analysis (DSA) Preboot Change History Version 10.2, Build ID $$BNU$$
Version 10.0, Build ID DSYLA1F o New System Support: x3650 M5 (5462) x3550 M5 (5463) NeXtScale nx360 M5 (5465) o New Operating System support: N/A o Problem(s) Fixed: N/A o Enhancement: N/A o Limitations: N/A o Special usage: - 2014 for M5 GA lnvgy_fw_dsa_dsala8b-10_2_anyos_anycpu_chg
HBA/LSI 12Gb SAS/SATA/SSD Driver for Windows 2012 R2/2012/2008 R2/2008
Version 2.51.01.00 - Moderate Impact/Low Probability of Occurrence ------------------ - Fixed issue where Disk Management would crash when more than one drive was connected. (SCGCQ00844533)
Version 2.50.92.00 - Moderate Impact/Low Probability of Occurrence ------------------ - Added generic timer facility in support of ATA Identify Hang and Microsoft Hung Commands issues (SCGCQ00755404) - Updated how SAS Port Attributes are handled to return only two states for each port: HBA_PORTSTATE_LINKDOWN or HBA_PORTSTATE_ONLINE. (SCGCQ00721232) - Fixed issue where adapter hangs on boot when number of attached enclosure slots exceeds the MaxTargets (SCGCQ00718669) - Corrected Initiator Port Addresses reported in PowerShell (SCGCQ00718669)
Version 2.50.75.00 - Initial release ------------------ lnvgy_dd_mpt3sas_2_51_01_00_windows_32-64_chg