Intel iMR 3108 ROC Hardware RAID Firmware 24.2.1-0021

Versions:

- Firmware 4.220.40-3267
- UEFI_Driver 0x06030501
- Hii 03.05.11.03
- BIOS 6.14.05.0
- Ctrl-R 5.03-0005
- BootBlock 3.02.00.00-0000
- NVDATA 3.1310.00-0070
- FCODE 4.16.08.00
- PCLI 1.07.05
- ROMENV 1.09

This download is valid for the product listed below:

- Intel RAID Controller RS3WC080

Fixes:

- Battery learn cycle continues to more than one cycle even FW has marked it as bad in previous reboot
- DIS mismatch while running system reboot script with dirty data in DDR
- Optimize onfi descriptor setting for offload
- FW crash while launching Ctrl+R
- Assertion failure in double DIP power cycle test case
- Elapsed time of Patrol Read in tty log is incorrect
- Elapsed time of Reconstruction is incorrect
- Firmware not taking "SMARTerEnabled" from NVDATA
- Rebuild triggered on an incompatible blocksize PD
- Upgrade 3108 MR FW failing
- PD erase does not work with SATA SSDs
- Diagnostics failed after Drives goes to shield state
- The DQlog.bin is getting corrupted after restarting the service of MRDiag
- protect the removal of the device when the device is still under initialization
- BIOS is not getting updated with the DISM progress
- Running out of free entries in task list due to recursive addition of tasks
- Wrong information of PD erase progress is shown by the firmware.
- DM_DISM_LOG_SENSE state is not present in DISM
- PI enabled VD is importing successfully even though "prevent PI Import" option is enabled
- Change the CRC filed size
- Fix the learn issue at powerup
- LSI_MR Controllers fails to set "FACTORY SETTINGS"
- VD's switch from Write Back with BBU to Write Through when an OCR is issued and BBU gets disabled.
- Fix scheduling of hostspare correctly
- seen DC error while running io's on 6.2 in Dev pull-push setup
- chaos fails on R6 VD after importing failed R5 VD and one of R1 CacheCade drive
- Defect fix for MSM does not show Create cachecade option
- To fix task list overflow issue on max config setup
- FW throws "MPT_LogDiscoveryErrors: SAS_TOPOLOGY_MAX_TOPO_PHYS_EXCEED" when connect 2 in multi-path
- Assertion failure in ../../raid/raidpci.c at line 10522
- Read VPD error message on 3108
- Configurable drives count in iMR reflected wrongly to applications
- wrong error code displayed and wrong error message for copyback command
- FW hits to MonTask: while running enclosure power cycle with 4K drives
- Issue with supercap energy calculations
- modify scap energy calculation
- directly attached SATA drive is not detected by controller
- Dual Context to Single Context switching does not toggle when Cobra expanders are cascaded with EDFB enabled PHY's.
- Defect fix for Montask observed while installing OS on JBOD
- Controller Hangs during execution of Storcli command ''/c0 delete termlog''
- Battery retention test failure during continuous loop test when there was power failure during erase
- Battery Enable (BBE) is asserted even when BBU/Supercap not present or even when all LDs are in WT mode
- TM timeout during enclosure pull while running heavy IO
- Learn is not completing at powerup
- 3TB Seagate PD size being reported as 2TB in Apps/Firmware
- Controller Reset due to Fatal Error with RAID6/60 configuration
- Elapsed Seconds displayed in firmware logs for background operations progress is improper
- Kill adapter observed while running IOs on JBODs and VDs
- CME handling of Invader
- Issue with Global Hot Spare assignment
- Drives connected through enclosure showing wrong slot numbers.
- Issue with MR 24.1.1-003 with JBOD Mode enabled
- (null) noticed while PD pull-push testing.
- SMPpassthru is not happening through storlib.
- System Reboots after issuing the OCR and kernel panic is observed
- on breaking multipath while drives are spinning up results in montask
- Initialize invalid interrupt as part of intInitialize function
- OEM server hangs after enabling EKM and rebooting.
- Associating cachecade VD to a secured VD from Storcli does not fail.
- Offload fails after corrupting BMT1 and BBT 2 address location
- Adding the log event corresponding to ER SCGCQ00493492
- BGI kicks in on 5R5 but shows no progress at any point of time.
- Defect fix for Unable create R0 in HII with 4K drives
- FW-FW hit 'MonTask: line 243 in file ../../raid/1078in' while issuing OS reboot command on 9340 controller
- CPLD input GPIO 21 driving when supercap is present and not present
- Pinned cache gets flushed even after memory battery problem occurs.
- VD erase restarts instead of resuming after reboot.
- Fix the cset issue
- Cache Memory Size of controller Information is ZERO
- Data Corruption observed while running FPIO on VDs; making-breaking multipath in overnight loop.
- Config loss after OCR during creation of config - when the drives are still spinning up
- Added an event to notify the user when command times out on PD
- FDE/SED drives does not reflect as FDE/SED drives.
- "Corrected medium error" event is logged for uncorrectable medium errors during PR
- To declare block bad at the earliest for the safer side the number of error correction bits shall be reduced
- Re-enable EKMS for OEM
- System reboots continuously after flashing Firmware in Solaris _11
- MonTask: line 7193 in file ../../bbu/onfi.c
- Kill adapter observed on large setup on enquiring Controller properties while drives spinning up from powersave.
- coverity tool reported issue 13545(Wrong sizeof argument)
- restore and flushing cache to PD are not reliable when 1st image of NF mem is corrupted
- Firmware is allow to create slices in cachecade drivegroup using SMIS utility.
- EKMS feature for OEM is disabled
- max temperature bytes read from VPD is should be swapped before using
- MR_6.2_DISM: FW hung due to DISM discovery failure on pulling the enclosure cable on reboot after running adpbattest.
- Controller alarm continues to beep after transitioning VD from degraded to optimal state.
- Handle OEM fault LED behavoir
- MR6.2_iMR_fury_DM_Issue_TUR
- In enclosure management page for DELL ,it shows some junk data in HII
- Controller Reset due to Fatal Error with RAID6/60 configuration
- CME logic using overlaid fields in c->rec
- Montask appears on running IO's on 64VD's while changing LD policies
- Enable OCR by default on GCA builds
- MR: on FW init, controller post adapter base port missing
- Invalid "Elapsed Seconds" value for FGI
- Rebuild and Consistency Check aren't progressed when it is executed by CTL+R
- Latest API changes in mr.h breaks FW compilation in mrdebug.c
- Build Issue using GNUmakefile for MR and iMR
- Storcli command to reset the boot ld on controller fails
- boot block compile fails because of an erroneous macro definition
- DATA TLB error faced while powercycle the enclosure during controller initialization
- foreign import fails if degraded VD is present in import
- Toolbox Console Text Display Tool
- Porting of the PR "spin up changes to accommodate multiple power domains simultaneously" to Invader.
- Validate the requested blocks for SCSI VERIFY commands on RAID volumes
- IOCInit multiple Reply Descriptor Post Queue addresses
- Firmware Change for Pre-boot management tool & boot capability for SPARC
- add reserved fields to IO Unit Page 7 for future use
- Disable thermal throttle changes
- Identifying the DDR, ONFI flash and SCap combination
- SGPIO back plane requires ramdon slot number support.
- Increase PCIe Receiver Margin
- Increase margin in PCIe Rx Serdes
- obsolete SAS IO Unit Control _TRANSMIT_PORT_SELECT_SIGNAL
- obsolete "Limit SATA max rate to 1.5 Gbps"
- Debug Queue Implementation and Support as part of enhanced MR Diagnostic support
- Increase the support for 32 slots
- Add Critical event in FW to indicate that an INIT operation was halted due to reboot or other failure
- Update code to modify GG and VPD at powerup
- Implement throttle mechanism to prevent thermal run away
- Change hash algorithm for cache row hash from a bit mask to a modulo
- DIS feature
- Signed images for FWDownload
- Don't interpret the cachecade BVD flags if the volume is created with old FW which does not support cachecade.
- Code-review action item for removing and updating names for AEN dism.
- Patrol Read - no progress
- WB Logical disks are moving to WT after corrupting BMT1,BBT1 and reboot.
- Deep Learn is not invoked in new battery, when the system is reboot before the first learn completes
- Transparent Learn Cycle is rescheduled even though it is not completed
- TFM error handling : Cache policy change for WB to WT if supercap is faulty
- Intermittent SEP errors on iMR and MR controllers
- When Cachecade LD is deleted invalid LD details is shown in the Cachecade size changed event.
- Hotspare Recovery Doesn't Start after Inserting Failed Drives
- FW sending wrong configuration data during boot msg pending state
- Upgrading from Software RAID to 9260-8i causes Alarm to sound
- ONFI_BBM Dev test build: FwTermLog does not preserve information before powerup.
- iMR TB Data loss when reconstructing from 1 drive R0 to 2 drive R1 with reboot
- incorrectly uses a slot offset value to enumerate HDD.
- In case of a certain OEM iMR shows R5 enabled.
- MSM log warning: Controller booted in headless mode with errors
- Set the Recovery mode page (WriteRetryCount and recoveryTimeLimit)
- If VD is secured while reconstruction is going on does not stay secured once reconstruction is done.
- SEAGATE hotspare drives remain spun up on LSI FW 12.12.0-0079
- Firmware is not updating recovery timer as intended
- Call DDF_discardInvalidPds only if DS3 is enabled
- LSI_MR Controllers fails to set "FACTORY SETTINGS"
- Controller Reset due to Fatal Error with RAID6/60 configuration
- VPD output of lspci –vvxxx showing [RV] Reserved: checksum bad, 86 byte(s) reserved
- Multi-stripe size in each raid level of R1,r5,r6 with SAS, DIF, 4K pinned cache offline vd import cause Montask
- Remove memory module on MR controller to downgrade to iMR does not work
- HII critical message form/Extra character "r" with configuration not supported critical message.
- Fw defect in PdStateSet()
- Controller Reset due to Fatal Error with RAID6/60 configuration
- Correct how the SCL high and low period values are calculated for the I2C to run at 400Khz
- For new spin up code get persistent devId for drives that are connected via internal back plane.
- Controller Reset due to Fatal Error with RAID6/60 configuration
- Drive security mode switching in MSM
- Unable to create another Virtual Drive when pulling first drive (degrading) in partial Virtual Drive where all drives are configured
- Baseboard Management Controller (BMC): STP passthru command failure
- External Key management security for iMR
- Re-keying a controller with External Key Management(EKM) security enabled, results in a corrupted key ID.
- Performing an online firmware update or Online Controller Reset(OCR) on a controller with External Key Management(EKM) enabled, results in all secured drives marked as foreign.
- MR 4K Validation : 4K VD assigned to 512K CacheCade
- Re-keying in External Key Management mode followed by disabling security and enabling LKM results in a bail out while creating a Virtual Drive
- VD goes offline when we start reconstruction on 32 Drive R0 VD
- Raise priority of pci config trap interrupt to critical
- Not seeing the correct Boot Message when the connection to the External Key Management server is down
- Switching to Local Key Management Mode(LKM) after a rekey in External Key Management(EKM) rekey without rebooting results in HII saying key incorrect but Firmware says key was created.
- On a controller with Local Key Management(LKM) enabled, user is asked to enter the security key to unlock and import a foreign, locked and External Key Management(EKM) enabled drive.
- FwTermLog does not preserve information before powerup.
- Implement I2C Slave Address for Sideband Management
- Defect Fix for VPD and inquiry callback issues in DISM
- spin up tracker dequeue is not handled for case in which TUR error happens even after MAX retries
- Too many unexpected sense AEN raised by CC during CME recovery on R5
- Too many unexpected sense AEN raised by CC during CME recovery on R5
- Fix GNU make file compilation
- Routines to inject bit error in NF and ecc error in DDR for ONFI
- Mismatch in LD id and target ID during set/reset boot device
- Mismatch in LD id and target ID during set/reset boot device
- Adding test routine for the bad block loop test for PR SCGCQ00423195
- SMART poll needs to issue log sense with page control set to current/ cumulative value
- FW breaks if wrong security key is entered
- on breaking the multipath in loop while drives are spinning up will result in montask
- FW hangs while creating 64 VDs on the controller with SLES11 SP2 PPC64
- While running reconstruction from R1 to 4 drive PRL11 with IO,FW hits to MonTask: line 1270 in file ../../raid/1078dm
- Static analysis tool defect
- Static Analysis Tool defects
- CLI Reporting of TMM VPD
- Rebuild triggered on an incompatible blocksize PD
- Issue with Global Hot Spare assignment
- Elapsed time of Reconstruction is incorrect
- Elapsed time of Patrol Read in tty log is incorrect
- "Corrected medium error" event is logged for uncorrectable medium errors during PR
- Increase margin in PCIe Rx Serdes
- Increase PCIe Receiver Margin
- Liberator Card loses its boot VD after many reboots
- Cache policy of Virtual Drives(VD) does not change to Write Through(WT) mode even after detecting a bad supercap.
- RAID configured on PD been erased with Normal or Thorough will lost after system reboot/reset
- Defect fix for kenel panic with iMR Fury with Free BSD driver
- Hitting montask line 1930 in file ../../bbu/onfi.c while running FP IO's on degraded VD's
- Cachecade configurations could not be created.
- Deep learn doesn't start when it is requested during relaxing.
- Write policy changed from WT to WB when the battery/Supercap is declared as bad
- Firmware crash seen while doing continuous sas cable pull
- Invalid "mode select(Page=0a)" issued during disk initialization phase.
- Defect fix for Montask observed while installing OS on JBOD
- Firmware returns asc value as 0 when failing IOs with medium errors
- display BMU fw version in battery design info
- Break Mirror for R10 should not be allowed
- Medium error correction event is not logged during CC
- 9240-4i on BCM RX67Q-6S motherboard is not getting required memory resources on boot
- foreign import fails if degraded VD is present in import
- L2/L3 cache error on reboot
- search for one wire devices prints are masked after the boot
- VPD info can't be retrieved for OEM enclosure
- If a BOOT Passphrase is enabled with SafeStor "MegaCli adpbios dsply a0" will display "BIOS will Stop on error" even if any other mode is set.
- Profile based VD creation - [Partial RAID options blocked until system reboot]
- Hard coded values should be removed in TFM module
- Fix repeated prints issue
- A data TLB error occurs when the External Key Management Server(EKMS) is down and drive security is changed from External Key Management(EKM) mode to Local Key Management(LKM) mode.
- HII & Ctrl-R: BGI operation still running on RAID0 VD even after PI protection disabled
- Getting a non informative message when trying to import a Virtual Drive after a re-key without reboot
- Deep learn will not start immediately when Relative State of Charge (RSOC) of battery is equal to 95%.
- Raid configuration loss after reboot - PDs in POWERSAVE
- Fix the learn issue at powerup
- Faulty protocol used for changing the Disk Write Cache Policy
- on breaking the multipath in loop while drives are spinning up will result in montask
- FW- FW hangs while creating 64 VDs on the controller with SLES11 SP2 PPC64
- Static Analysis Tool defects
- Static Analysis Tool defects
- Enabling DQ support bit in FW
- The restore successful and unsuccessful prints to be handled properly
- not enough bs found during double Dip power cycle Test
- Montask in divert.c seen when injecting 64K consecutive errors on R0 VD, Patrol Read running in background.
- Single image support for ONFI
- Assertion failure (not enough BS found) in double DIP power cycle test case
- DC observed upon performing OCR on a degraded R1 (AWB).
- 9260-8i(SAS2108) hitting OOB SRAT timeout for a long time causing OOB commands to fail
- modify scap energy calculation
- wrong error code displayed and wrong error message for copyback command
- Bad or Missing TMM boot message is changed to Critical error
- 9260-16i's drive locate/fault LED header doesn't work
- MR Controller does OCR due to Temp Buffer overflow
- The Firmware does not change the Virtual Drives's from Write Through to Write Back after a battery manual learn cycle is complete
- SLES 11.3 x64 Halts when Loading Kernel During Warm Reboot Stress
- Initialize invalid interrupt as part of intInitialize function
- Switching the security mode from External Key Management(EKM) to Local Key Management(LKM) on iMR controller will result in a corrupted controller key.
- SMART poll needs to issue log sense with page control set to current/ cumulative value
- Static Analysis Tool defects
- Power removal of enclosure FW hits to MonTask: line 4056 in file ../../bbu/onfiBBM.c
- Multiple interposer drives will not be detected by MegaRAID firmware.
- Infinite retry happens on Test Unit command when HDD didn't respond to TU command during the initialization.
- Secondary RAID key data page read failure 6 seen on some controllers and key vault combination
- StorCLI_F showed firmware download failed but really took
- Assertion failure in ../../raid/raidpci.c at line 10522
- FW Crash when using SCSI Passthrough SCSI_READ_10(0x28) to read 512 bytes on a RAID5 Native 4K drive.
- When pinned cache present VD property not changing from WB to WT.
- PDs go online without VD created & App doesn't allow any VD creation on passing a wrong parameter in Spanned VD.
- Customer is hitting an ECC error
- Invader: Cache Memory Size of controller Information is ZERO
- Incorrect variable type in MR Read/Modify/Write path
- StorCLI_F showed firmware download failed but really took
- cache restore failure message displayed though cache restore was successful
- System hanged after pulling out drives from a 8Raid-6 array during OCR in Debian OS
- Change SES Green/Amber LED behaviour for OEM ACHILLES JBOD enclosure.
- FW breaks if wrong security key is entered
- While executing path-failover test with an SATA-SSD drive behind an interposer, the device is marked failed.
- Card Running iMR Machine Checks when Requesting Host Memory
- iMR FW update caused card to FW fault state and card cannot boot to OS.
- system black screen during S1 Sleep test
- Firmware Crash during Controller StartUp when large number of drives are connected
- Block IO tests from SCT IHV test suite failed
- Disable Debug print in UEFI Driver
- Not able to boot into BIOS boot menu
- In IsValidHandle() - Free the buffer if status is not EFI success for LocateHandle
- Memory Leak issue in UEFI driver when reconnect happens
- Remove private GUID and use EFI_KMS_PROTOCOL_GUID defined in UEFI 2.3.1 spec
- Disable Debug print in UEFI Driver
- UEFI hang when SATA is connected
- UEFI hang when SATA is connected
- UEFI Driver should calculate checksum before setting bios data (MR_BIOS_DATA)
- Uninstall HII CAP and Driver Health formset if platform calls Start again without calling GetHealthStatus
- 9240/9260/9270 UEFI driver issue with RD640/RD440 platforms when user having 2TB or 4TB VD
- 9240/9260/9270 UEFI driver issue with RD640/RD440 platforms when user having 2TB or 4TB VD
- Follow UEFI Spec on interpreting In and Out Transfer Length that upper layer sends
- Block IO tests from SCT IHV test suite failed
- Follow UEFI Spec on interpreting In and Out Transfer Length that upper layer sends
- Allow user to edit VD size (that is free form input) as it is cumbersome to modify the size using +/-
- Save and Restore Configuration – MR
- Transition a Thunderbolt MR controller to run in iMR mode = Controller fail in HII After Memory removed
- When controller is in safe mode the PD menu displays inaccurate text and hangs when clicking a PD
- HII\EKM-LKM without reboot should show proper message
- HII cannot assign partial space less than 1 GB
- fixing code review defects (AIM version + UCM GUID + deallocation)
- Profile Based Raid1 Creation fails as invalid arguments
- Naming inconsistency in option present in Create VD and Create VD - Advanced
- condition checking is not right
- Allows user to select EKM mode when supportEKM is set to 0
- Using 84 PD encl when clicking on HW, Advanced then Enclosure Management, it completely freezes the system
- Allow user to select RAID level when controller has other VD + error handling
- Enter security key for locked drives screen, press down to get to OK, OK will show up again on right
- unable to show UF003 as well the first two forms + XML special character handling
- fixing memory leak of 128 drive enclosure implementation
- Able to select EKM even when the EKM FOD key is removed from the system.
- Drop down option to set phy linkspeed tp 12 Gbps in HII browser is not available.
- Save and Restore Configuration - MR w/ Premium Features (SED and Cachecade)
- Issues when trying to change EKM security key from change security key option.
- EKMS is down but it is still possible to select drive security in MSM and HII during creation.
- Going from LKM to EKM asks for a Current Security key Identifier and Security key.
- Update the latest API change in HEFU
- UCM Phase 1: Typo "Clear Configuration" -> "Clear Configuration"
- EXPAND operation fails in HII for prl11
- Streamline the usage of max PD in limited configuration path (alias UCM)
- Help content for Free space in Select Drive group is inappropriate
- UCM Simple Configuration: Allow Clear Config always + Suppress Checkbox (Uf002.PDs) when VD count is >0
- added a while loop and take care of the multiple PD Groups scenario.
- AIM (Agentless Inventory Manager) Issues
- Typos in help content for profile VD for Write Policy
- Profile based VD creation path incorrectly showing protect VD/secure VD checkbox when user clicks on drive selection criteria
- config using 8PDs/16 partial VDs and all stripe sizes and HSPs does not restore
- After creating a VD with UCM tool the manual configuration PD section still shows configured PDs as UG, and duplicates
- Controller not seen in HII Browsercan after switching from EFI WebBIOS
- Checking the Enclosure Management settings in HII causes the system to reboot.
- Request to support 128 drives wihin single enclosure
- UCM Phase 2 support
- (Implemented) - Create a formset with classGUID that customer has provided.
- Request to change the help content for write policy in create VD form
- Obtain the Physical Slot number from the PCI Bridge to which the Controller is connected.
- System hang observed when the "PdInfo.userDataBlockSize" is set to 0
- HII cannot assign partial space less than 1 GB
- RSOD during POST after enabling SecureBoot - HEFU.ROM should report exact size of .EFI image
- Typos in help content for profile VD for Write Policy
- HII cannot create a VD using unconfigured good drives as long as PD locate is set on the drives
- Obtain the Physical Slot number from the PCI Bridge to which the Controller is connected.

- Getting memory conflict during boot
- Memory access issue when loading Ctrl R
- OOB function corrupts EBDA area used by other server function
- not allocating memory correctly when MR and iMR are in the same system
- Optimize reading IO port in INT15h path
- System reboots during IO because FW didn't receive permanent host memory from OpROM
- OpROM is displaying incorrect value for the number of INT13h device it supports
- Provide an option to limit the max number of INT 13h devices presented to system BIOS
- Check the controller status bit in the function which hooks INT15
- Allocate only two 64k
- Add boundary checking when allocating EBDA and PMM memory for BIOS post.
- OpROM is displaying incorrect value for the number of INT13h device it supports
- Memory access issue for Mercury when loading Ctrl R
- Memory allocation: Allocate only two 64k
- OOB function corrupts EBDA area used by other server function
- Change Emergency Spare to NONE with enable emergency on Smarter checked will fail with error=03
- ctrl-r dosent list all the drives to do replace drive.
- Power savings button not enabled on Ctrl-R
- ctrl-H doesn't list the drives to do replace drive while doing copyback in PD MGmt Screen
- Selecting "Manage Link Speed" in CTL+R causes Fatal Error with SAS 9343-4i
- Implemented the safe mode in Ctrl-R
- Unable to make R10 through Ctrl R using 4k drives
- Default (Initial) setting of VD doesn't reflect MFC's setting
- /c0/pall show and Linkspeed information is not in same in Ctrl-R.
- Changes in BIOS mode (pause on error) to be in sync
- Need to display controller memory in Control-R
- showing "BIOS Mode" in both the pages of ctrl mgmt
- Help content is not updated with the Key stroke added
- VD Erase will be disabled in Ctrl+R when Initialization on particular VD
- Write policy is not added with writeback with BBu when MFC is set
- iMR boot Block code initial changes
- Errbios load fails on Fury adapters
- Merged 6.0 boot block changes to 6.1
- Bootblock is using NVRAM data even if NVRAM is uninitialized
- Configure chip reset to link down
- Added the code to enable OCM RMW errors
- Coherency bit is disabled in SRAM TLBs
- Numerous changes for specific customers
- PCLI(1.07.05) Sparc: Command "cli set securitykey=L
- Fcode CLI support for Invader (MR only)
- Define and use capability bit to identify the StorCLI