Open muellpi opened 3 years ago
I have the same issue on a system with TXE 4.x Gemini Lake, I managed to use FIT to decompress the TXE region and ran me_cleaner against that region and seems to work.
now how do I inject the TXE region into the full dump rom ?
Hello, recently I had success cleaning some older intel fw systems and namely a newer intel I5-8250 desktop cpu, all with ME firmware and external SOIC flashing. But I had no success when appying me_cleaner to a TXE 4.x fw image from a N4100 celeron cpu. With surely active TXE image me_cleaner gave me this output:
After debugging me_cleaner.py a bit, I found, that it was not able to read or assemble the correct me_start and me_end region from fw header.
While flreq[0] and flreg[1] seem to converge to some reasonable values, namely file descriptor start..end = 0..4096 bytes Bios 4096..6652k me bytes seem to be odd.
`In [1]: me_start Out[1]: 134213632
In [2]: me_end Out[2]: 4096 ` I am not sure if the fw layout changed and flreg[2] is not the right place to look for me region or if
return (flreg & 0x7fff) << 12, (flreg >> 4 & 0x7fff000 | 0xfff) + 1
is no more the right way to decode flreg[2].I have two questions :
Could someone give me a cue how to help me_cleaner to find my me region and allow it to proceed in wipe process ?
could somebody explain me what these operations calculate ?
return (flreg & 0x7fff) << 12, (flreg >> 4 & 0x7fff000 | 0xfff) + 1
In the meantime I managed to soft disable ME with coreboot ifdtool, but it would be nice to remove.
Some additional information:
hexeditor output of my fw header relevant first part:
okteta GD25LQ64_org.rom
Relevant 6 bytes for flreq :
0000:0060 | FF 7F 00 00 FF FF | ..ÿ.ÿ...ÿÿÿÿÿÿÿÿ
complete header:
some ipython debugging
Output of MEAnalyser python tool:
Finally Output from coreboot ifdtool: