IBM M1015 Experiences

[pete_c]

mobilenvidia,

I haven't posted in a while and have finished building my base Asus E35 / M1015 cross flashed to IR mode and it is working.

Only issue is that I cannot get to the set up page on the M1015. I saw it once. Is there some other key combo to get to the setup page?

Or should I go with another mini-itx board?

Patrick,

Yes the case was too expensive (answering your pm from a while ago).

 

[mobilenvidia]

At boot press the key combo that allows you to choose a boot device.
Then at the LSI splash screen, press the key combo to access the BIOS (CTRL+C ?)
Once in the boot selection page, choose the LSI controller
This should then lead to the LSI BIOS page.

 

[pete_c]

Thank you mobilenvidia. I can now see the LSL configuration.

I have enough space/SATA ports in the case now to add SSD drives to the mix.

 

[Dimahnbloe]

Sorry to bother you guys/gals, but Iv'e run out of options and am hoping that someone here has experienced this and will offer some help.

Iv'e got an m1015 cross flashed to a mega-raid 9211 and I am currently dealing with the windows server 2008r2 "system hang on shut down or restart" issue that I can't seem to solve. It was reported as being a fixed issue in one of the earlier driver change logs. What's happening is that they system functions perfectly until the need to either shut down the system, reboot, or hibernate. Any attempt to do any of the three seems to end with a blank screen after the "Windows is shutting down message". The change log on the driver version where I believe the issue is identified indicates that this is due to "Power SRB handling"

When this occurs, there is no exception messages, no blue screens, and no issues reported in the event log. It's almost as if the system believes that it shut down properly but the m1015 is keeping the power on. I think Iv'e tried just about everything to solve this matter, except for the one thing that would probably be the most helpful: using older drivers.

My system is a quad socket Tyan s4985 fully loaded with 4 quadcore processors. It is based on the nforce 4 professional chipset from nvidia and has 2 pci-e 16x slots, 2 4x slots, and one standard pci slot. Ive got 32gigs of ram, 8gigs per cpu bank. I am using the m1015 to boot a single Samsung 830 series 256 gb drive. Nothing fancy. I had no issues with my old megaraid pcie controller on this same motherboard

Here are the things Iv'e tried:

1. Swaping m1015 to every single slot
2. Also tried isolating the m1015 by removing all other cards from their slots. then repeating step one
3. Disabling any onboard hard disk, ethernet, com, and usb controllers in cmos.
4. Disabling option roms.
5. Eliminating all memory down to 1stick per cpu
6. Elminating all cpus except for cpu 0
7. Installing server 2008r2 on different hard disks
8. Flashing to IT mode with and without bios
9. Flashing to IR MOde
10. Using it as an M1015 with no cross flashing
11. Flashing multiple m1015 firmwares
12. Changing moptherboard chipset drivers

I am hoping that someone can help by either offering advise on how to fix this based on experience, or maybe lend a hand by pointing to the location of older drivers that I can download. Would be even cooler someone could upload some old 9211 driver packages on a reply post. I am not to sure how many versions that were released over the years, but The more packages that can be uploaded the better.

Thanks for the help in advance.

 

[mobilenvidia]

When the M1015 is in its native mode (iMR) or LSI9240 MegaRAID mode this is completely different from the Card being in LSI9211 IT/HBA mode.
Basically the card is a RAID controller in one mode and an HBA in the other (chalk and cheese)
Each mode requires completely different drivers.
If you are getting issues in either mode, I'm pretty safe in saying that the issue is not the card, but most likely a Power setting on the Mobo.

 

[mobilenvidia]

Added

P15 IT and IR Firmware, BIOS, UEFI installer,UEFI_BSD, SAS2flash/SAS2flsh installers

See first post for link

 

[Aluminum]

So not completely related, but I have discovered my lenovo X230 + expresscard tool refuses to finish posting if the board is plugged in. Tried everything, timer delay, forced Gen1, Legacy/non-UEFI mode, etc, everything set the same as my X220T.

I also tried hotplug in windows and it detects the card as RAID controller just fine (LSI 9240 going by hardware ids) but hotplug after freedos is loaded and megarec never sees the card

Sucks because my X220T works great but I'm selling it and the only new thing with X230 is the chipset: QM67-->QM77

 

[mobilenvidia]

IBM Latest FW 20.10.1-0107

ServeRAID M1000 Series SAS/SATA Controller Firmware Update

Adapters Supported: ServeRAID M1115 SAS/SATA Controller for IBM System X (81Y4448)
ServeRAID M1015 SAS/SATA Controller (46M0831)


Version 20.10.1-0107 - Suggested
-------------------
Fixes:
- Fixed grammatical errors in WebBIOS.
- Fixed an issue in WebBIOS Drive Group Definition screen where enclosures
should not be selectable in the Drives column.
- Fixed an issue in uEFI-WebBIOS where the white bars do not match up under
VD properties page.
- Fixed an HII browser issue where a drive performing a copyback displayed
the state as "Replacing" but other utilities showed the state as copyback.
- Fixed an HII browser display issue where SATA drives showed "0" for the
value of RPM under drive properties.
- Fixed a display issue where the HDD port number showed incorrectly in an
HII browser.
- Fixed an HII browser issues where creating a global hotspare gave the
user an error message.
- Fixed an HII browser issue when security is changed with local key
management the changing security report current mode is disabled as
external keu management.
- Fixed an HII browser issue where the user got the following error message,
"The request command has invalid arguments," when changing the security
key report.
- Fixed an HII browser issue where the user was unable to save the
configuration after adding an extra span during array creation.
- Fixed an HII browser issue where the return value was not correct when
the browser called extract config with a result of null.

 

[9jack9]

I'm currently using a M1015 flashed to P15 IT mode, however, within unraid, I keep getting errors like these in my syslog.

Jan  2 06:41:15 Tower kernel: sd 0:0:2:0: attempting task abort! scmd(f0aa0480)
Jan  2 06:41:15 Tower kernel: sd 0:0:2:0: [sde] CDB: cdb[0]=0x28: 28 00 00 16 30 18 00 04 00 00
Jan  2 06:41:15 Tower kernel: scsi target0:0:2: handle(0x000c), sas_address(0x5001e6739eae7fe6), phy(6)
Jan  2 06:41:15 Tower kernel: scsi target0:0:2: enclosure_logical_id(0x5001e6739eae7fff), slot(6)
Jan  2 06:41:15 Tower kernel: sd 0:0:2:0: task abort: SUCCESS scmd(f0aa0480)

I've tried different versions of the firmware, I've tried not using an ivy bridge processor, but nothing seems to be making a difference :/

Since unRaid just wants the disks in JBOD configuration, I've left it in IT mode, but perhaps I should go utilize IR mode? Suggestions would be greatly appreciated.

 

[nitrobass24]

How do I tell if my card is IBM or LSI?

 

[tchuyev]

I finally managed to flash my M1015 to LSI 9211-8i in IR mode using my Asus P9X79 WS, the same motherboard that give me headaches in the native DOS prompt.
I was able to flash it via the UEFI shell. The method I used it`s a simple one and I think it should work for everyone who had the "Failed to initialize PAL" error.


You can read below all the steps I`ve made, starting in DOS and finishing via the EFI Shell:
1. I downloaded the latest updates (sas2008.zip) for LSI 9211-8i from mobilenvidia`s laptopvideo2go.com forum: http://forums.laptopvideo2go.com/topic/29059-sas2008-lsi92409211-firmware-files/page__hl__m1015. I was a bit lazy and I found very convenient to download everything in one go. Thanks mobilenvidia!
2. Then using a USB stick I created a MS-DOS Bootable Flash Drive. You can finf how to do it here: http://www.sevenforums.com/tutorials/46707-ms-dos-bootable-flash-drive-create.html
3. After that I extracted the content from sas2008.zip on the USB stick.
4. I set up in Bios to boot from the USB and in DOS prompt, I followed the instructions from the first post in the earlier mentioned thread:

Convert LSI9240(IBM M1015) to a LSI9211-IR mode
Type in the following exactly:
megarec -writesbr 0 sbrempty.bin
megarec -cleanflash 0
<reboot, back to USB stick>

5. In DOS, after reboot using "sas2flsh -o -f 2118ir.bin -b mptsas2.rom" command I got the infamous "ERROR: Failed to initialize PAL. Exiting program".
6. I went to ArchWiki, to the Unified Extensible Firmware Interface (UEFI) page.
The web address is: https://wiki.archlinux.org/index.php/Unified_Extensible_Firmware_Interface#UEFI_Shell
Scroll down to "UEFI Shell " and download "x86_64 UEFI Shell 2.0 (Beta)" and "x86_64 UEFI Shell 1.0 (Old)"
To flash my card I used "x86_64 UEFI Shell 1.0 (Old)".
I renamed the "Shell_Full.efi" to "shellx64.efi" and copied on the USB stick, over the previous files from the "sas2008.zip" archive.
7. I also had to copy on the stick the sas2flash.efi program. You can download the "Installer_P13_for_UEFI " from the LSI support page: http://www.lsi.com/channel/products/storagecomponents/Pages/LSISAS9211-8i.aspx
8. Attach the USB, restart the PC and go in the UEFI Bios. Check if the UEFI USB drive is your first boot option. On the top right corner click on the "Exit" button. You will get a new menu and the last option is "Launch EFI Shell from filesystem device" Click on it and you will go in the UEFI Shell.
Here I was able to continue the flashing procedure:

sas2flash.efi -o -f 2118ir.bin -b mptsas2.rom
sas2flash.efi -o -sasadd 500605bxxxxxxxxx (x= numbers for SAS address)

9. Done.


I thought it`ll be a good idea to post in detail all the steps, especially for guys like me, who don`t do everyday things like this.

Also, a big thanks to Servethehome.com for the great article "IBM ServeRAID M1015: Performance of the LSI 9220-8i". That was the article that pointed me to the M1015. Also, thanks mobilenvidia for putting all together. It helped me a lot.
Now I`m happy, I have a little monster ready to chew all the data I`m going to feed him.
Cheers,

Hi, I've followed the step-by-step procedure above and still can't get up and running with my M1015 (SAS9220-8i).
The message 'Error: Failed to initialize PAL. Exiting Program' still appears.
> Downloaded 'sas2008.zip' from page 1.
> Tried to update most binaires with all latest versions from LSI.
> Tried to boot from USB stick as per procedure above, but also from CD-ROM with FreeDOS.
> Tried on 2nd different PCs with X79 and Z77 chipsets.
Still no luck, grrrrr...

The only thing I've managed to do so far is to execute both 'megarec' commands to respectively flush and erase the PCIe card's configuration.
But the latter is no longer detected by DOS or Windows.

What do you recommend I should do now?
Is the M1015 SAS9220-8i NOT supported?
In the worst case scenario, how can I get back to initial IBM M1015 configuration?
Thanks ;-)

 

[mobilenvidia]

In the worst case scenario, how can I get back to initial IBM M1015 configuration?
Thanks ;-)

http://www.servethehome.com/ibm-serveraid-m1015-part-4/

 

[mobilenvidia]

LSI FW 20.10.1-0126

Bug Fixes and Enhancements:
===========================
Firmware defects:
SCGCQ00262426 - If VD is secured while reconstruction is going on does not stay secured once reconstruction is done.
SCGCQ00313883 - Battery continually stating Battery Yearly Deep Discharge Relearn Pending. Please Initiate a manual Relearn
SCGCQ00319537 - System hang on IBM machine with a critical boot message
SCGCQ00327006 - Event (MR_EVT_CTRL_RESETNOW_DONE ) is not generated after OFU is completed
SCGCQ00329851 - EEPROM corruption(BQ2060 errata) due to corrupted value of design Capacity written in FCC for iBBU07
SCGCQ00332181 - INVALID SGL causes dump into montask, OCR does not recover. MonTask: line 377 in file ../../dm/mpt2/mptrec.c & MonTask: line
SCGCQ00333679 - reverting the changes of SCGCQ00295604
SCGCQ00333696 - EEPROM corruption(BQ2060 errata) due to corrupted value of design Capacity written in FCC for iBBU07
SCGCQ00334032 - Code review checkin
SCGCQ00335607 - Firmware upgrade/downgrade not possible due to missing env stamping
SCGCQ00336716 - Megamon issue when disabling the drive security on a MR controller with spun down FDE drives.
SCGCQ00337894 - OFU is not happening when tried to upgrade firmware.
SCGCQ00338008 - Flash of package with OFU request
SCGCQ00345497 - Code review: With the new approach for CtrlOFUPossibleSet/Get() shutdown path broken
SCGCQ00346275 - Copyback on a drive hits kill adapter
SCGCQ00347253 - Firmware fails to return proper value when flashed same firmware image with different stamping with OFU request.
SCGCQ00353344 - Data Corruption during Raid Level Migration
SCGCQ00335935 - Coverity Defect fixes
SCGCQ00286152 - GetConnectorIndex() Out of bound array access
SCGCQ00334889 - Multiple switch cases are not handles correctly in debug functions
SCGCQ00314781 - Controller always enter HSM when it is set to HCOE
SCGCQ00317767 - Code Review Defect: Correct the error in copy back path smart errors path
SCGCQ00333346 - Dont interpret the cachecade BVD flags if the volume is created with old FW which does not support cachecade.
SCGCQ00340772 - Address Action Item SCGCQ00340495 (Check coding standard for structure OfuRequestState)
SCGCQ00332658 - Out-of-bounds access (OVERRUN_STATIC) of an array "ar" in pdCopyBackSpinupCallBack
SCGCQ00332651 - Out-of-bounds access (OVERRUN_STATIC) of an array "nvInfo->lockKeyBlob.keyCheck" in displayNvKeyInfo
SCGCQ00329070 - Coverity Defect
SCGCQ00315029 - PD clear operation is not handle correctly with spun drive drives
SCGCQ00332588 - Fix coverity defects 10216 and 10425 (overrun and potential NULL pointer access)
SCGCQ00307050 - Not able to flash both 9265-8i and 9285-8e card with 5.3 patch 1 firmware.However, it is ok on 9266-8i.
SCGCQ00353322 - Unable to enable ocr using application due to FW ocr setting taking precedence
SCGCQ00328109 - Out of bound access in LDBBMListInit
SCGCQ00337060 - Coverity Defects
SCGCQ00341791 - Support for mutlipath topology is broken
SCGCQ00330344 - FW halts with a a kill adaptor on performing an onffline of the drive under copyback. line 1188 in file ../../raid/1078dma.c
SCGCQ00332099 - Out-of-bounds access (OVERRUN_STATIC) of an array "ar" in pdDcmdRebuildCB

Firmware Enhancements:
SCGCQ00302600 - Added more debug information
SCGCQ00307949 - True FW Online update for Liberator
SCGCQ00307950 - BGI is in endless loop continuously restarting after completion
SCGCQ00308988 - Improve foreign config import for iMR with mixed JBOD/RAID settings
SCGCQ00314258 - Need the ability to determine what disk a logical LBA associate with on the fly, api call/algorithim/etc. on Liberator
SCGCQ00323446 - Add NVDATA support to make 512 Emulated drive support as configurable for customers.
SCGCQ00345203 - megacli reports battery pack is missing on hot insertion
SCGCQ00314965 - iBBU07 - SOH bad will be declared for low capacity battery before starting learn cycle
SCGCQ00307956 - iBBU07: EEPROM and register values will be updated to revised values at first boot after FW flash.

BIOS:
Version change only

UEFI Driver:
SCGCQ00262858 Defect Debug message is shown from EFI app choice prompt
SCGCQ00308870 Defect EFI Driver 518\When disconnecting the driver in EFI, I am still able to exectue the EFI utility menu.
SCGCQ00312683 Defect UEFI Driver not timing out in 3 mins when FW is in undefined state
SCGCQ00320619 Defect UEFI: Build Device Path and Get Next Target function of SCSI Pass Thru is not returning correct EFI status for certain cases
SCGCQ00335617 Defect Drives are reporting timeout on Write Buffer cmd
SCGCQ00339371 Defect UEFI Driver fails SCT Test
SCGCQ00343869 Defect Enhance critical boot message handling
SCGCQ00338066 EnhancementRequest UEFI - HAMR: Fire LD list query based on maxHANodes
SCGCQ00338782 EnhancementRequest If SetOptions is called for Invader dont show EFI WebBIOS and EFI CLI option
SCGCQ00340581 EnhancementRequest UEFI Driver should check pMRCtrlInfo ->maxHANodes with > 1 to determine HAMR

HII:
SCGCQ00318010 (DFCT) - Allows to select Cachecade VD as Boot device
SCGCQ00319005 (DFCT) - HII will not allow config to be cleared no drives attached
SCGCQ00319010 (DFCT) - HII while in safe mode locks up when entering View drive
SCGCQ00293056 (ENHREQ) - HII: LSIP200198414/LSIP200229508/LSIP200231613: Disables all requirement for human input at boot time
SCGCQ00317934 (ENHREQ) - Add MFI State error string for the new MFI stat error codes
SCGCQ00317991 (ENHREQ) - Remove 'Adaptive Read Ahead' option from every MegaRAID management utility (PR: SCGCQ00295908)

 

[mobilenvidia]

Windows driver 6.505.5.0

Bug Fixes/Enhancements:
=======================
Changes from version 6.505.4.0 to 6.505.5.0
1)SCGCQ00349835 Windows MegaRAID driver issue on 9260/9280 cards when using CacheCade

Changes from version 6.505.3.0 to 6.505.4.0
1)SCGCQ00342417 Performance drop in 256 KB seq. Write by driver 6.505.2.0 comparing to driver 5.2.127

Changes from version 6.505.2.0 to 6.505.3.0
1)SCGCQ00339795 Modify WinOS INF file to use the proper product name for OEM
2)SCGCQ00339786 9260/9280/9240 Format Drive(VD) does not complete succefully in Windows

Changes from version 6.505.1.0 to 6.505.2.0
1)SCGCQ00330474-Invader : System fails to wake up from m S3 sleep.
2)SCGCQ00330224-Invader : System fails to wake up from Hibernate.
3)SCGCQ00329951-WHQL:Hibernation related test cases fails.
4)SCGCQ00330366 WHQL:Crashdump support test fails.

Changes from version 6.505.0.0 to 6.505.1.0
1)SCGCQ00319076 Driver failed to install(9280) controller as primary

Changes from version 6.504.3.0 to 6.505.0.0
1)SCGCQ00319409 4k Sector Windows
2)SCGCQ00280021 Support Windows WDCFG util for Windows MR driver

MegaRAID storage Manager 12.10.03.01

 

[mobilenvidia]

IBM FW 20.10.1-0107

ServeRAID M1000 Series SAS/SATA Controller Firmware Update

Adapters Supported: ServeRAID M1115 SAS/SATA Controller for IBM System X (81Y4448)
ServeRAID M1015 SAS/SATA Controller (46M0831)


Version 20.10.1-0107 - Suggested
-------------------
Fixes:
- Fixed grammatical errors in WebBIOS.
- Fixed an issue in WebBIOS Drive Group Definition screen where enclosures
should not be selectable in the Drives column.
- Fixed an issue in uEFI-WebBIOS where the white bars do not match up under
VD properties page.
- Fixed an HII browser issue where a drive performing a copyback displayed
the state as "Replacing" but other utilities showed the state as copyback.
- Fixed an HII browser display issue where SATA drives showed "0" for the
value of RPM under drive properties.
- Fixed a display issue where the HDD port number showed incorrectly in an
HII browser.
- Fixed an HII browser issues where creating a global hotspare gave the
user an error message.
- Fixed an HII browser issue when security is changed with local key
management the changing security report current mode is disabled as
external keu management.
- Fixed an HII browser issue where the user got the following error message,
"The request command has invalid arguments," when changing the security
key report.
- Fixed an HII browser issue where the user was unable to save the
configuration after adding an extra span during array creation.
- Fixed an HII browser issue where the return value was not correct when
the browser called extract config with a result of null.

 

[Davie]

damn! that's a whole heap of bug fixes - it might fix all of my ZFS problems with Seagate 3TB drives in my Norco backplanes (or it might not!)

 

[mobilenvidia]

Those 2x FW updates are for the M1015 in its native LSI9220/9240 mode not IT mode in which you should be running ZFS

 

[mrkrad]

Could you take the eeprom off the M5000 PERFORMANCE key, read it, write it to the M1000 advance key ? Just for backup purposes.

 

[mobilenvidia]

It's a 6pin IC, only 2 pins are used, so it's a serial arrangement.
I think to program it you need to remove it and there will probably be another pin you use to do so.
To erase it, I think you need a UV light.

But no point, my M5000 key works on the IBM M1015 as a M1000 key
I wonder if a M1000 key does the same on a IBM M5015/5014 ?

 

[cactus]

You can hook to it on the PCB. Use small grabber clamps or solder to the exposed legs/pads. I am surprised no one has tried to roll their own, one should be able to capture the output while it is running and emulate with a similar flash or uC. Only two wires connected makes more sense for some kind of 1-wire EEPROM. LSI card sends it a command, chip regurgitates a key.

Edit: Infact Dallas has a 1-wire 6pin TSOC 256bit EEPROM.