AMI Aptio 2.17.1255 header parser bug in A64

[StackerDEV]

I came across this playing with old junk..

ROM Bin dumped:
https://mega.nz/file/JYZkCKyY#4wGqzIvyKimRUhpG2vsB-2CCrirfdaasylJrvxpNRZg

Problem description:
Extract as is and replace body is working good in UEFITool 0.28.0 for above rom. I did the mod, eeprom and system posting/booting properly no issues.
When I tried in A64 header bugged (corrupted), can't replace anymore I flash rom sys no post. I have no H80 display connected yet on backorder, trying to make a SFF system from junk mb. Any other options I have not messed with using AMICBP for improvering thermal tripoints and unlocking more options for full life extension.. All I did search BMP magic change bootlogo..... Should be super easy effortlessly but somehow not possible with this rom in the new version A64, try ir yourself.

Details of what I am trying to do.
->>

Machine
Rom name: Terra-Clevo-Medion_1530_i3-6100U.rom
PCB revision: MA50LP_v1.0 SKL PCH-LP Mobile(U) 21/C1 Premium SKU PCH RC 1.9.0.0.
AMI Aptio ver: 2.17.1255
Aptio core ver:

Info:

File GUID: 4B1E6486-6C0E-538E-1357-8FB97063DFD6
Type: 02h
Attributes: 00h
Full size: 23BFh (9151)
Header size: 18h (24)
Body size: 23A7h (9127)
State: F8h
Header checksum: 34h
Data checksum: AAh

RAW SECTION

Extract as is ..

header + bmp data
D2 E7 3D 19 & 42 4D CE ********

HEADER

FILE MAGIC data after RAW section dump:
42 4D CE
BM
CS5 header no good, magic must be 42 4D CE not 42 4D B8 or no post.

Terra-Clevo-Medion_1530_i3-6100U_PCB_MARK_MA50LP_v1.0.rom.zip

[NikolajSchlej]

It's very unclear from this report what the issue is, sorry.
I did found a bug in that "Extract body..." doesn't work for some reason and that will be fixed in A65 for sure, but I need a better explanation to understand what's up.

[NikolajSchlej]

The fact that A64 can not replace is known, as the editing part is still not done in the new engine, tracked by #67.

[NikolajSchlej]

The section itself looks fine and has the expect BPM image magic

[NikolajSchlej]

I've extracted the body of that raw section in 0.28.0 and a fixed version of A65 (62d96a1), and the files are identical.

% shasum -b 0280.raw 
f37d66c72b3ce3664541771adfe36b817ce547b1 *0280.raw
% shasum -b A65.raw 
f37d66c72b3ce3664541771adfe36b817ce547b1 *A65.raw

[NikolajSchlej]

Closing this as "can not reproduce", please reopen it if you do not agree.

[StackerDEV]

That seems odd (I remember at Alpha 64 there was a odd situation header was corrupted when I write back and look again with UEFITool it was that way around). Next time I try to make it reproduceable, but I was aware of possible error still. Project needs more reversing and that takes time and lots of testing.

I might have made an error at some point in the process.
Thanks for investigating and then amazing project. Will try to support it if possible.