Asus P7H57D-V EVO with 2 onboard Marvell Controllers
Motherboard BIOS VERSION: 1903 DATE: 09/28/12

From my boot screen I see the following:
Marvell 88SE91xx BIOS Version 1.0.0.1025
Marvell 61xx BIOS Version 1.1.0.L73

If I disable the controller in the BIOS upgrade says no chip found

Leaving the controller enabled it will upgrade with message firmware upgraded.
Before rebooting I check to see if firmware was upgraded the incoming and current firmware are the same.

After rebooting the Motherboard BIOS reflashes the Marvell 88SE91xx back to older firmware.

Need help to add the upgraded Marvell 88SE91xx & 61xx to my Ausus P7H57D-V EVO Motherboard BIOS VERSION: 1903.

I have attached copy 1903.rom & MMtool326 in the P7H57Bios.zip

Win 10.0.17763 Build 17763
Intel(R) Core(TM) i5 CPU 750 @ 2.67GHz, 2668 Mhz, 4 Core(s), 4 Logical Processor(s)
BIOS Version/Date American Megatrends Inc. 1903, 9/28/2012
SMBIOS Version 2.6
Embedded Controller Version 255.255
BIOS Mode Legacy

Current Driver Marvell 91xx 1.2.0.1049
Current Driver Marvell 61xx 1.2.0.8400

So you need BIOS mod, to disable the in-BIOS Marvell updating? (* I checked, this is not possible in this BIOS) Also, when you reboot, only the 9123 is downgraded again, correct? The 6111 remains as you updated it?
The 9123 may be auto-recovering that FW due to what you flashed in is not correct? What FW are you updating it with? Please link source so I can read details there, if any
Stock rom for 9123 has 9120 ID, so if you put in other than that ID, that may be why it's auto-recovering it. FYI, the 6111 rom in BIOS has 6121 ID

Fernando says here, for 91xx, unless BIOS offers RAID for this controller it's pointless to update the rom - which this BIOS does not offer. Also, if you are using SSD, this will not fix the issues << my 2 cents
AHCI & RAID ROM Modules

So, are you sure you want to try mod BIOS with updated 9120 Marvell roms?

I have attached the FW i'm using to update the 91xx. I also included a Flash.txt that shows B4flash, After Flash, After Reboot.
I have not been able to find a FW update for the 61xx.

Sorry, I guess I read you wrong, I thought you updated the 61xx and it was the only one that updated OK and remained updated.

There's the problem, in your txt file, this package has been modified/is not for your chip ID.
Image = 1.0.0.1813
BIOS = 1.0.0.1038 (ID 9123) << Wrong ID for what's used in your current BIOS FW
Autoload = 1.0.0.1802
Firmware = 2.3.0.1007 (20-09-2012 11:02:34)
mvf_mag = 1.0.0.2461 - modded for cFeon EN25F40 chip << Not your system

Also, some onboard motherboard controllers like you have can't easily be updated with package like this one.
Did you get this on a forum where someone with same or similar motherboard (Same general series chipset P55/P67/H67 etc) used it and showed update success?

I'll make you mod BIOS with both updated

I can't remember where I got the FW update it years ago. I tried several others this one seemed to work the best even though the bios changed it back. After reading about all the problems with the Marvell Chip and others having similar problems I gave up. Never had any luck with 61xx.

Win 10 seemed to work better with the Marvell Chip so I decided to try again.

I want to Thank You for your help.

Work best??? How is that, when it's failing? That doesn't make sense But yes, I do know what you mean, it acted like it was trying at least

You're welcome! I'll do in BIOS change out on both for you, working on it now. Give me a few more minutes (15 maybe)

As for the Marvell chip itself, and how it functions (especially with SSD), this isn't going to matter about OS used it will still stall, stutter etc,
These 91xx chips (9120/9123+9128) were the first 6Gb/s controllers (actual 5Gb/s) and just can't handle actual SSD speeds without coughing up dust

# Edit - Sorry, I cannot safely perform this update either, unless you have flash programmer to recover? The Marvell bootloader module in your BIOS does not even contain Marvell ID or the device ID, so I don't feel confident updating it with the update one, I think it would fail.
Additionally, I think if I left that in place, and updated the other two components, that would cause fail to due to the bootloader not also being updated. If I just update the AHCI/RAID ROM module itself, that would probably fail due to above thoughts as well, but maybe this could be tested.
However, this is all my guessing, it may be all fine to update, but I wouldn't try it without a flash programmer and confirmed OK backup made.
Please wait, lets see what Fernando says about this when he has time, I know he's dealt with this a lot. I can update 6121 rom for now if you want, it looks like a proper safe update
On 9120, I'll read through this mess thread too, while we wait - [Discussion] Marvell 91xx/92xx SATA3 Controller BIOS modules
And this one - [Discussion] Firmware Update of the Marvell 91xx SATA Controller
If you see anyone asking about similar things I mentioned below, let me know, in case I don't get to them in these threads before you do
* Edit, looks like here, this is suggested against for your chip anyway, due to no RAID - [Discussion] Marvell 91xx/92xx SATA3 Controller BIOS modules (12)
What I mentioned about the AHCI Rom update only is suggested as safe too, so if you want we can try that only.

@Fernando - I'm sure you've ran into this before, how was it handled, did updates per usual fail when bootloader (E6) didn't contain Marvell ID/Device ID and was then updated with proper Marvell files that did?
Same question on E7 module, neither in stock BIOS look anything like the update modules. E7 stock is 3KB even, vs 200KB update Additionally, autoload looks split between two more modules (E8-E9)
The only thing that looks normal here is the actual AHCI ROM from CMSCore, can that be updated by itself?