@TimeWalker - check lenovo's module: 17772369-D262-4B90-9F31-BDC41F2663A5_1_759.ROM
It's an ME update, but an old one (7.0.0.x) - looks like it's an executable though?
11-11-2012, 05:00 PM (This post was last modified: 11-11-2012, 05:08 PM by TimeWalker.)
Downloaded an Intel® Server Board S1200KP BIOS KPC2060H and there's indeed
ME 7.1.52.1176 - C43791FC-E05B-4AA0-84B1-F14547885C70_0_21.ROM 1.27 Mb
ME 8.0.13.1502 - 098D0689-4245-4F65-80C9-7F3202C5F44E_0_28.ROM 1.48 Mb
inside the bios capsule ...
@jkbuha, yeah it appears to be ME 7.0.0.0054 .. which explains why the size is so tiny. But this one is actually referenced as MEBx .. which my explain why it starts with an executable header (MZ)
Posts: 48
Threads: 1
Joined: Dec 2011
Reputation:
11
@TimeWalker
You said earlier on that you have Clover working. Does that mean you've got your touchpad working in Clover too?
btw, I've uplaoded the new version of my tool. It should extract/inject everything correctly. Make sure though since I compiled it as 32bit this time as it's more stable.
11-11-2012, 05:37 PM (This post was last modified: 11-11-2012, 05:43 PM by TimeWalker.)
hmmm... will try doing something else tomorrow. Thanks to @dmazar's tricks I was able to mount the firmware's file system!
And there are certainly more modules loaded up then I see being unpacked from the capsule.. also there are two file system which I'm not sure what the fsnt0 is ..
a simple ls in the filesystem's root reveals the following: 243 modules (i'm thinking of passing all of the *.efi modules to an external drive to see if ME fw is there .. and some other stuff possibly?)
and I tried running the Internal Shell from the firmware ..
As you can see from the image below- we are still stuck with UEFI version 2.0 (whilst most modern boards are running 2.31 already)
A quick dh shows that shell now has a handle of 1AE ..
If the filesystem was always mounted like that we would be able to do: bcfg boot addh 1 1AE "Shell 2.0" -opt 0x40000000 0x0015
This would make a new Boot menu entry called "Shell 2.0" and would call for the respective handle's GUID from the fw when key 0x0015 (F11) is pressed. But since I can only mount the fs from a third-party driver .. we have no internal shell ... for now.
Also I quickly checked if the ME from X1 can be run as an application.. nah-uh, it can't.
Posts: 523
Threads: 0
Joined: Aug 2011
Reputation:
23
11-11-2012, 05:52 PM (This post was last modified: 11-11-2012, 06:16 PM by TimeWalker.)
0xF and 0x0F is the same thing you can't split a byte in hex
here's the module after applying the patchset.
and these were probably my edits ... lemme see ..
yeah they were lol.
The power management patch unlocks write access to MSR Register 0xE2
nebster Wrote:I'm assuming this is what unlocks all the options in Tiano BIOSes
Correct.
nebster Wrote:Now this I have no idea, was it just a test?
This fills the manufacturing unicode string with spaces. The string reads as follows: "Below is reserved for RD, not DELLOIL request."
You can actually see what it says by decoding HEX-> ASCII
For new VBIOS an entire module has to be reintegrated.
There was also a CPU microcode update but it's to big to patch it like that... besides it sometimes differs in size and you have to know what was the binary data for the previous version, so automating this is pretty pointless if you ask me.
I have just one suggestion @nebster ...
don't call the modified file .WPH.mod .. it's annoying to having to go to Folder Options ans removing 'hide extensions for known file types' because VLC threats .mod as a media container ... and I hate Midnight Commander (or TC) -like apps.
Posts: 48
Threads: 1
Joined: Dec 2011
Reputation:
11
11-11-2012, 07:32 PM (This post was last modified: 11-11-2012, 07:42 PM by nebster.)
@TimeWalker
As per your request, I've modified it so that it now asks for a filename as well.
Also, I have implemented patch versioning, compression, checksums, comments and lots of checks just in case the patches become corrupt.
I'll upload a patch in a bit as an example. At some point, I'll add embedding binary files into the patches so we can replace whole files if we want. Maybe I'll even implement a way of copying data from the old file into the new one if it is useful for anything.
------------------
Ooops, forgot to hit "Post Reply"