@ fedy:
Welcome at Win-EAID Forum!
Which case do you mean resp. to whome did you address your tip?
Regards
Fernando
Tried to update another Supermicro BIOS
Xeon-Mainboard: X10SLM±LN4F
x10slmp4_424.zip
Replacing of Modules works fine,
but when trying to insert newest CPU-Microcode, I get into a loop (allways have to enter “1”).
This BIOS is 6 files GUID 17088572-377F-44EF-8F4E-B09FFF46A070. And I do not know which files need to be changed for correct operation of the BIOS. So my advice to make the CPU microcode update manually or contact technical support or leave as is.
Realtek PXE and RPL ROM 2.61
Realtek PCIe GBE Family Controller Series v2.61 (01/07/15)
RTEGROM.261.zip (188 KB)
@lordkag
um… i mailed asrock…
" i have another question…
until very recently i kown thought there is no EFI driver for marvel 9172 sata cipset…
now i know Marvell EFI drivers are already out there…
and some of mineboard surpport boot with Marvel 9172 in Disable CSM mode…
so has asrock any plan to update now BIOS contain EFI marvel sata driver??"
and asrock sent me this…
"
Dear Sir,
Thank you for inquiry.
We suggest user use the PCH UEFI for your system.
If there any concern to use the PCH UEFI to your system, please feel free to contact us.
Thanks for your mail.
Warmest Regards
ASRock TSD"
what is PCH UEFI mean???
PCH is Peripheral Controller Hub aka chipset, they are saying that you should use chipset SATA ports for non-CSM UEFI-bootable drives.
Found Broadcom UNDI 16.4.6 in sp70087. Also, it turns out ValleyView also has a long version, like VBIOS 7.1.1007, GOP 7.1.1003, GOP 7.2.1008.
Broadcom Lan UNDI 16.4.6.rar (84 KB)
X79PFS_3.zip (5.24 MB)
@lordkag
few day’s ago asrock gave me new bios. 3.10A
thay say 3.10A bios surpport marvall cantroller EFI boot
but it is not work to me…
in my system i can see driver loaded and EFI Shell scan file System…
but strangely only UEFI mode EFI shell can not read contents of file system…
you can see the fs2: in ‘update3.jpg’ the same file system ‘EFI for windows’
in legacy mode… EFI SHELL can read all file and bootloader and file
i report it asrock… and asrock said
'We tested the BIOS P3.10A in our side before release to you.
It can install the Windows in UEFI mode(AHCI) and boot normally.'
do you know why this bug occured?
edit:
i attached asrock 3.10 bios
it centain marvell RAID driver and ahci driver…
it is almost same you gave me few days ago but guid is diffrent
AMITSE and marvel 9172 DXE replaced new one…
@dygkset
What is the problem exactly? Is UEFI booting not working at all for Marvell controller, no HDD/SSD detection, or just the UEFI Shell has problems? Either way, it is Asrock business to fix their problems. The only thing I recommend is to run this beta file in UBU and update the Marvell EFI AHCI driver to 1.1.0.1015. Other than that, nothing can be done from our side, without sources and EFI development knowledge.
the problem is can not boot target disk cennceted marvel sata…
becouse uefi driver can detect HDD/SSD and scan file system well but can not read in it include boot loader…
the reason i ask to you not asrock that asrock said thay also test it… and they can boot and there is no such a bug exist.
do you think the test of asrock incorrent?
edit:
forget it
i found why diffrent resert ocured betwen asrock’s test nad my test…
m1 m0 port do not work…
m2 m3 port works…fine…
@SoniX
A new release of the script:
- added extraction and version for ASIX USB UNDI and only extraction for SMSC USB UNDI, Realtek USB UNDI. Also x86 detection.
- Intel ME or TXE detection, along with variant and date. Thanks to Igor and Plutomaniac.
- added Intel RST RAID for PEIM, found in Insyde UEFI.
- added CloverView and ValleyView detection + extraction. I had to write the version detector in python, since it has a structure like 7.x.yyyy and is also x86 or x64 coded.
- usage of uefifind to clear false positives. Unfortunately, on my machine it was too slow to use uefifind for everything.
- added many new GUIDs, but most of them for Insyde.
- added AMD firmware SMU and IMC extraction, by reading coreboot and SMU vulnerability. If it has a header, it will also print version and platform: TN - Trinity, RL- Richland, KB - Kabini, KV - Kaveri, ML - Mullins. If not, only version. It seems SMU is OEM independent, for the same platform and version you will always get the same firmware. Not sure about IMC, as I only have 3 samples from different platforms.
- Broadcom NIC firmware detection. Can’t find a sample to also add extraction.
- separated ASM104x in ASM1042 and ASM1042A.
- added ASMedia, NEC firmware extraction
- a better rename for GOP addon, with platform_version_VBT template.
- in UEFIExtract.py using regex for clipboard reading, so you can copy a big chunk of data and only the GUID will be read.
- a few fixes in UEFIExtract.py
- added an option for official AMD microcodes found in Linux repositories. Use “acpu amd” for official bin.
- moved oromext3 into oromext. Now, using oromext will extract all OROMs from file, be it in CSMCORE, ffs or no header. Multiple images in OROM will be extracted as one image.
- kept oromext3 as an alternative, but added an extra argument. Using “oromext3 all” will extract individual OROMs, even when they are linked, like in SCSI or VBIOS+GOP.
- added “romscan” function by reading PnP spec. This takes an OROM and displays all available information from PCI structure. Using “romscan pnp” will also display PnP info.
- added a “diff” option. I wanted a quick way of inspecting those non-empty paddings. This extracts only those blocks that are different than scan byte.
- namelink improved with many IDs.
- fixed GUID detection for OROM. It should work in most cases. Haven’t found one that fails.
- about x86 detection. For UNDI is simple, since they are usually built on top of UDK. We have “(0xFFFFFFFF - (UINTN)” for x86 and “(0xFFFFFFFFFFFFFFFFULL - (UINTN)” for x64. For ValleyView, I read the PE header. 4D5A is the start of efi, at start + 3C is the pointer to PE header. PE begins with “50 45 00 00” and then code type follows: 014C for x86, 8664 for x64
- added version for all Intel UNDI, like 10G or 40G.
#Extractor.rar (59.1 KB)
@lordkag
Pay attention to the BIOSes of Supermicro (C60x chipsets). The modules OROM are located in GUID 9F3A0016-AE55-4288-829D-D22FD344C347. New MMTool 5 allows you to work without problems.
@ lordkag:
It is great, what you have published within your last post. This will let SoniX do another big step forward with his UBU tool.
Thank you very much, my friend! 
I think we noticed this before in EVGA SR-X W888_008. But at that time I thought it was a mistake from EVGA. But the new MMTool does indeed reads this file for OROMs. I added this file for OROM search and also fixed oromlink to read the links. If the main pattern is not found, I just locate the first OROM and apply the first pattern from oromext to find the start of links. The rest is the same.
I noticed this new MMTool can replace the old one for both Aptio4 and AptioV. So I made some test changes, to use MMTool5 for all AMI EFI. Unfortunately, it provides mixed results. It works for the compressed volumes in Acer Iconia and Toshiba Encore and EVGA X99, but MMTool5 fails for Tiano in Thinkpad and also for LZMA in Intel releases from ME pack, like "Intel ME Tools 10.0.25.1048_utility". Also the false positives might rise, since with UEFIExtract I could just silence them.
It seems AMI has no fail-safe, it wants only standard AMI images. So I still keep UEFIExtract as a backup. UEFIExtract is also the only way for Insyde, Phoenix SCT or any strange EFI. UEFITool engine is just better than MMTool, it is like CodeRush beat them at their own game. Even PhoenixTool feels better constructed than MMTool, despite having a different purpose.
Here are the test changes. To be used only with MMTool5.
#TestMM5.rar (51.5 KB)
Will there likely still be updates for the OROMs and microcode of my board?
Anyone?
Updates will be, if there are new versions of OROM and microcode
Work has begun on a new version UBU 1.10.0, is partial support for the BIOS on the platform AptioV.
For 6-series most probably not. And you don’t need them if everything is working as it should. Let’s not exaggerate.
Not exaggerating at all. The tool is available to keep things updated. If there’s a new release of the OROM’s for series 6 boards then might as well use it. It’s just an easy update anyway. My question was really about asking if there will likely be new OROM upgrades in the future, not if I need them. Most people tend to keep things updated even if they don’t need to, because we can.