LSI 9270-8i IT mode

[vanfawx]

I'm going to give flashing a 9260-8i a try today.

@w0mbl3 It's a retail 9270-8i. My 9260's are retail as well as far as I know. And yes, the card shows up as a 2308 chip.

07:00.0 Serial Attached SCSI controller: LSI Logic / Symbios Logic SAS2308 PCI-Express Fusion-MPT SAS-2 (rev 05)
        Subsystem: LSI Logic / Symbios Logic 9207-8i SAS2.1 HBA
        Physical Slot: 2
        Control: I/O+ Mem+ BusMaster+ SpecCycle- MemWINV- VGASnoop- ParErr+ Stepping- SERR- FastB2B- DisINTx+
        Status: Cap+ 66MHz- UDF- FastB2B- ParErr- DEVSEL=fast >TAbort- <TAbort- <MAbort- >SERR- <PERR- INTx-
        Latency: 0, Cache Line Size: 64 bytes
        Interrupt: pin A routed to IRQ 24
        Region 0: I/O ports at 5000 [size=256]
        Region 1: Memory at fbff0000 (64-bit, non-prefetchable) [size=64K]
        Region 3: Memory at fbf80000 (64-bit, non-prefetchable) [size=256K]
        [virtual] Expansion ROM at e7100000 [disabled] [size=1M]
        Capabilities: [50] Power Management version 3
                Flags: PMEClk- DSI- D1+ D2+ AuxCurrent=0mA PME(D0-,D1-,D2-,D3hot-,D3cold-)
                Status: D0 NoSoftRst+ PME-Enable- DSel=0 DScale=0 PME-
        Capabilities: [68] Express (v2) Endpoint, MSI 00
                DevCap: MaxPayload 4096 bytes, PhantFunc 0, Latency L0s <64ns, L1 <1us
                        ExtTag+ AttnBtn- AttnInd- PwrInd- RBE+ FLReset+
                DevCtl: Report errors: Correctable+ Non-Fatal+ Fatal+ Unsupported+
                        RlxdOrd+ ExtTag+ PhantFunc- AuxPwr- NoSnoop+ FLReset-
                        MaxPayload 256 bytes, MaxReadReq 4096 bytes
                DevSta: CorrErr+ UncorrErr- FatalErr- UnsuppReq+ AuxPwr- TransPend-
                LnkCap: Port #0, Speed 8GT/s, Width x8, ASPM L0s, Exit Latency L0s <64ns, L1 <1us
                        ClockPM- Surprise- LLActRep- BwNot-
                LnkCtl: ASPM Disabled; RCB 64 bytes Disabled- CommClk+
                        ExtSynch- ClockPM- AutWidDis- BWInt- AutBWInt-
                LnkSta: Speed 5GT/s, Width x8, TrErr- Train- SlotClk+ DLActive- BWMgmt- ABWMgmt-
                DevCap2: Completion Timeout: Range BC, TimeoutDis+, LTR-, OBFF Not Supported
                DevCtl2: Completion Timeout: 50us to 50ms, TimeoutDis-, LTR-, OBFF Disabled
                LnkCtl2: Target Link Speed: 8GT/s, EnterCompliance- SpeedDis-
                         Transmit Margin: Normal Operating Range, EnterModifiedCompliance- ComplianceSOS-
                         Compliance De-emphasis: -6dB
                LnkSta2: Current De-emphasis Level: -6dB, EqualizationComplete-, EqualizationPhase1-
                         EqualizationPhase2-, EqualizationPhase3-, LinkEqualizationRequest-
        Capabilities: [d0] Vital Product Data
                Unknown small resource type 00, will not decode more.
        Capabilities: [a8] MSI: Enable- Count=1/1 Maskable- 64bit+
                Address: 0000000000000000  Data: 0000
        Capabilities: [c0] MSI-X: Enable+ Count=16 Masked-
                Vector table: BAR=1 offset=0000e000
                PBA: BAR=1 offset=0000f000
        Capabilities: [100 v2] Advanced Error Reporting
                UESta:  DLP- SDES- TLP- FCP- CmpltTO- CmpltAbrt- UnxCmplt- RxOF- MalfTLP- ECRC- UnsupReq+ ACSViol-
                UEMsk:  DLP- SDES- TLP- FCP- CmpltTO- CmpltAbrt- UnxCmplt- RxOF- MalfTLP- ECRC- UnsupReq+ ACSViol-
                UESvrt: DLP- SDES+ TLP- FCP- CmpltTO- CmpltAbrt- UnxCmplt- RxOF- MalfTLP- ECRC- UnsupReq- ACSViol-
                CESta:  RxErr- BadTLP- BadDLLP- Rollover- Timeout- NonFatalErr+
                CEMsk:  RxErr- BadTLP- BadDLLP- Rollover- Timeout- NonFatalErr-
                AERCap: First Error Pointer: 00, GenCap- CGenEn- ChkCap- ChkEn-
        Capabilities: [1e0 v1] #19
        Capabilities: [1c0 v1] Power Budgeting <?>
        Capabilities: [190 v1] #16
        Capabilities: [148 v1] Alternative Routing-ID Interpretation (ARI)
                ARICap: MFVC- ACS-, Next Function: 0
                ARICtl: MFVC- ACS-, Function Group: 0
        Kernel driver in use: mpt2sas

So it full on thinks it's a 9207-8i.

 

[vanfawx]

@nthu9280 here's a screen shot of the sas2flash -list after flashing the card to the 9207-8i firmware, but before I re-assigned the SAS address.

 

[vanfawx]

So the 9260-8i did not show up to sas2flash in dos after I wiped the flash, unlike the 9270-8i that shows up as a 2308. No big deal though as I was able to re-flash the 9260-8i as I kept the SBR and SPD for the card before I attempted it.

The 9270-8i that thinks it's a 9207-8i works perfectly. I hooked it up to a SAS expander and tossed some drives on. No issues. Tried both ports, again, no issues picking up the expander in the C1100 I put it in.

So this works, but so far only with the 9270-8i.

 

[w0mbl3]

Which version of sas2flsh.exe did you try with the 9260? I've found different versions work with different cards.. it might be worth trying something designed to work with a PCIe 2.0 card like the 9210-8i, from a really old firmware version (e.g. sas2flsh.exe from P12 firmware or earlier). Here:

SAS 9210-8i Host Bus Adapter

Your 2208 reports like mine does - shows up as a 2308_2(D1) stepping now. My actual 2308 cards are E1 stepping (HP H220), confirmed by prying off a heatsink and checking last night (as they're sitting on a shelf, not in a pc). -listall output from my reflashed LSI2208 below

sas2flash -listall -l foo.txt
Adapter Selected is a LSI SAS: SAS2308_2(D1)

Num Ctlr FW Ver NVDATA x86-BIOS PCI Addr
----------------------------------------------------------------------------

0 SAS2308_2(D1) 20.00.07.00 14.01.00.06 07.39.02.00 00:01:00:00

Finished Processing Commands Successfully.
Exiting SAS2Flash.

 

[vanfawx]

I always use the P14 sas2flsh.exe on DOS. I've used it to flash 9205-8e cards with no problems. Funny enough, it's a HP221 that was labeled as a 9207-8e, but ended up actually being a 9205-8e.

I'll give the P12 sas2flsh.exe a shot. I ended up flashing the 9260 back to its MR firmware, but I can easily wipe it again.

Thanks again for your notes on this @w0mbl3 It's been a very interesting learning process!

 

[nthu9280]

I always use the P14 sas2flsh.exe on DOS. I've used it to flash 9205-8e cards with no problems. Funny enough, it's a HP221 that was labeled as a 9207-8e, but ended up actually being a 9205-8e.

I'll give the P12 sas2flsh.exe a shot. I ended up flashing the 9260 back to its MR firmware, but I can easily wipe it again.

Thanks again for your notes on this @w0mbl3 It's been a very interesting learning process!

I'd imagine H221 would show the older stepping of 2308 (C0 Ithink)

 

[vanfawx]

I'd imagine H221 would show the older stepping of 2308 (C0 Ithink)

You are correct! And it requires the 9205-8e firmware (PCIe 2.0). The funny part is it was sold as a H220, is labeled as a 9207-8e, but ended up being a H221/9205-8e.

Anyhow, I think the 9260-8i (2108) is a dead end for converting into a HBA directly. I might poke at it more tomorrow if things are quiet.

 

[BLinux]

You are correct! And it requires the 9205-8e firmware (PCIe 2.0). The funny part is it was sold as a H220, is labeled as a 9207-8e, but ended up being a H221/9205-8e.

Anyhow, I think the 9260-8i (2108) is a dead end for converting into a HBA directly. I might poke at it more tomorrow if things are quiet.

just curious if you ever got a chance to try the 2108 conversion to HBA? and if so, any success?

 

[vanfawx]

just curious if you ever got a chance to try the 2108 conversion to HBA? and if so, any success?

Hey, I tried, and failed. This only works with 2208 RoC products. :-(

Luckily though I was able to put the RAID firmware back on the 2108 card, so I didn't brick it

 

[nthu9280]

Glad to hear you were able to restore the flash.

I didn't see a strong reason other than academic excercise or you have stash of 2208. The HP h220 / h221 can be obtained for $45 range pretty regularly and the 2108/2208 based cards are typically more expensive.

Speaking of this I need list my H700/2108 that I bought a while back before I knew what the heck I was doing with enterprise HW and STH forums.

Sent from my Nexus 6 using Tapatalk

 

[vanfawx]

The 9270 I converted never really worked properly as a raid card. It was on the junk pile to be thrown out, so it was an easy candidate. But as a converted HBA, it works great.

 

[w0mbl3]

Another data point.. IBM M5110 (2208 card) apparently flashes using 2308 firmware too...

Comment yesterday from Nathan on my write-up:
Flashing LSI 2208 with IT firmware to use as an HBA - My Wired House

And tweet confirming..
Nathan Buck on Twitter

 

[vanfawx]

I wouldn't be surprised if this works on most 2308 based cards. It's a nice option to have for sure!

 

[svtkobra7]

Hate to necro an old thread, but I found what I believe to be a curious side effect of this process - perhaps analogous to the buzzer that someone else report: that being the "SAS Fault" LED on my motherboard SMCI X9DRH-7TF is apparently hip to the game regarding my shenanigans and decided to light up red. Is this of any concern?

Also, it came with a BTR-0022L-LSI00279 BBU, which I used to see reference to status, i.e. charging, etc. during boot. Without the MegaRaid firmware present any longer, does it serve any use other than taking up space?

My approach => followed @ w0mbl3 's guide Flashing LSI 2208 with IT firmware to use as an HBA - My Wired House. I made it through the process (without error), so I'm assuming I didn't screw up, subsequently enabled pass through for my "2308" in ESXi, and then added it to my FreeNAS VM, which is chugging along just fine.

Finally, popped my

(firmware flashing that is).

Boot

sas2flash - FreeNAS

# sas2flash -list
LSI Corporation SAS2 Flash Utility
Version 16.00.00.00 (2013.03.01)
Copyright (c) 2008-2013 LSI Corporation. All rights reserved

        Adapter Selected is a LSI SAS: SAS2308_2(D1)

        Controller Number              : 0
        Controller                     : SAS2308_2(D1)
        PCI Address                    : 00:1b:00:00
        SAS Address                    : 5003048-0-1914-7701
        NVDATA Version (Default)       : 14.01.00.06
        NVDATA Version (Persistent)    : 14.01.00.06
        Firmware Product ID            : 0x2214 (IT)
        Firmware Version               : 20.00.07.00
        NVDATA Vendor                  : LSI
        NVDATA Product ID              : SAS9207-8i
        BIOS Version                   : 07.39.02.00
        UEFI BSD Version               : N/A
        FCODE Version                  : N/A
        Board Name                     : SAS9207-8i
        Board Assembly                 : N/A
        Board Tracer Number            : N/A

        Finished Processing Commands Successfully.
        Exiting SAS2Flash.

 

[vanfawx]

@svtkobra7 Hey! The LED should be harmless. And the battery as well should be useless as you don't have the MegaRAID firmware to interface with it.

It's neat to see more people getting use out of this process!

 

[svtkobra7]

]Hey!

• Howdy! Thanks very much for the reply.

The LED should be harmless. And the battery as well should be useless as you don't have the MegaRAID firmware to interface with it.

• I would think the "SAS FAULT" to be harmless as well. It literally is just an LED, that you would never see unless you the top popped on the chassis. But for someone who has to map CPU1 fan to FAN1 header, and CPU2 fan to FAN2 header, i.e. OCD, it is mildly annoying. But that annoyance is offset by the benefits.
  
• I wonder how it knows? (rhetorical, not bothering you with that question, as I'm sure I could research).

It's neat to see more people getting use out of this process!

• Agreed; thank you for all of your assistance regarding this genre. I read quite a few of your posts last night and you definitely know what you are doing.
• Personally, with my luck / willingness to experiment, I'm surprised everything went as planned without any fires!
• My primary driver for going down this path was the desire to use an integrated HBA (of which the 2208 is not) in my new servers. Of course that doesn't work for everybody, but it does in my case. While I'm 5+ years late to the 10 GigE game, I really like the idea of a mobo with that component (not to mention 10 GigE) integrated. So I'm glad all worked out, in no small part to you.

"Looking forward" / speaking out loud, although I wouldn't have a single clue how to to do so, it would be fun / interesting to "benchmark" a 2208 v 2308. Dual 800 MHz v Single, respectively, and in my case 1GB cache v. none. I would *think* such a ROC, converted to IT mode HBA, would have tangible performance benefits.

 

[vanfawx]

Unfortunately the 1GB cache, like the battery, is not used when you're running the HBA firmware. It'll work and perform just the same as another 2308 based card. And the fault LED probably expects a certain response from the RoC that it no longer can send as it's not running the proper firmware. It's not wrong, since the card is not working as it was designed ;-)

Thanks again to @w0mbl3 for figuring this out and documenting it so well!

 

[svtkobra7]

Thanks again to @w0mbl3 for figuring this out and documenting it so well!

• @w0mbl3 +1 on that - thank you.

Unfortunately the 1GB cache, like the battery, is not used when you're running the HBA firmware. It'll work and perform just the same as another 2308 based card. And the fault LED probably expects a certain response from the RoC that it no longer can send as it's not running the proper firmware. It's not wrong, since the card is not working as it was designed ;-)

• Gotcha - I buy the non-use of cache straightaway, and I'm not debating you, but what I'm hearing (as an analogy) is that the ECU is "reprogrammed" to shut down an entire cylinder bank of a V8, which at first is tough for me to conceptualize but I suppose the PPC architecture actually lends itself well to bypassing a processor? Boo.
• What about the 2 MB of cache on the 2208 block diagram (lacking on the 2308)?
• So it sounds like my supposition is not just incorrect, but it couldn't be more incorrect and I have it backwards, i.e. [mobo with integrated 2208 flashed to 2308] < [mobo + 2308 card] from a performance perspective.
• Perhaps I missed the boat entirely, not that I care to save a PCIe slot (I have 5 available after 2 are used by NVMe), but perhaps I would be better off throwing the 9211 back in there (not that I like that idea). I find all of this quite interesting and took the pre-flashed HBA for "granted" and thanks again for your posts.

 

[vanfawx]

I'd keep the 2308 over a 2008 based card like the 9211. And no worries, I'm glad people find my posts helpful!

 

[kennyyang]

Unfortunately the 1GB cache, like the battery, is not used when you're running the HBA firmware. It'll work and perform just the same as another 2308 based card. And the fault LED probably expects a certain response from the RoC that it no longer can send as it's not running the proper firmware. It's not wrong, since the card is not working as it was designed ;-)

Thanks again to @w0mbl3 for figuring this out and documenting it so well!

So does the speaker

I flashed my IBM M5110 to 9207-8i recently, and had the same speaker issue. Gonna remove it as what you did.