Not sure, SpeculationControl 1.0.8 does not report about CVE-2018-3640 (yet)
CPU Micro-Code Rev. 25 for Haswell CPU works without problems.
UBU v1.70 b6 Dev - Test mCode Update
If you have the opportunity to test the updating of microcodes, then write down your results.
Example:
- Model, version BIOS
- Update: OK or ERROR
- After flashing: Work or Brick
* Alternative, if previously brick.
*-Update with MMTool: Ok or Error
*-After flashing: Work or Brick
If you have a brick after flashing, then specify a link to the BIOS file.
Asus Rampage VI Extreme
Model : 1401
Update : OK
After flashing : Brick
It’s ok if the code is modified
http://dlcdnet.asus.com/pub/ASUS/mb/LGA2…E-ASUS-1401.zip
My result:
UBU v1.70 b6
- Asus Z87M-Plus 1107
- Update mCode: Ok
- After flahimg: Work
@SoniX Yes
I’ll test
I was ok with this : [Discussion] UBU Tool related Questions, Reports and Suggestions (270)
And UBU 1.70 b5
@SoniX
I test tonight or tomorrow morning (depending on the time I go home) with your modded bios
EDIT : you can put back the link of the modded bios
How do you update the LAN OROM/EFI on the latest dev builds? It shows that there is a newer version but when I try clicking update it just says, "temporarily not available."
@tistou77
Ok.
Try to flash this BIOS. I have already replaced the microcode, I just have to flash it.
http://rgho.st/7VZbCRlMk
Report the result. If brick, then what POST code.
Thanks.
@SoniX
Just to let you know.
UBU is crashing, at least for me, since version b5. The problem is that the last one of the ")" at the line 240 of the "UBU_dev.bat" file in version b6 must be removed.
In short: Line 240 is ")))" and must be "))" for UBU to stop crashing.
@Phoenix48
Line 210
if %amd%==0 goto trix
to
2
rem if %amd%==0
goto trix
Add:
I do not yet test BIOS for AMD, so I forgot about these lines.
I reuploaded the archive "*_fix1".
I see that a new development version of the UBU Tool is available. When time permits, I will have fun testing it this week by updating older BIOS images to versions in which only the microcode changed. That way, I can compare the CRCs of the modified BIOS images to the CRCs of the BIOS images which I have already downloaded from various motherboard vendor web sites. My idea is to verify whether or not the latest development version of UBU Tool has resolved the offset issue when updating CPU microcodes. I will be sure to track all of my efforts so that if any issues are encountered, then I can pass on the information to the programmers for their review.
@SoniX
tested your mod bios and “brick”, code 55
EDIT: I also tested with the b6, same result
It’s good with the b5 and this code change
2
3
4
5
6
echo [Replacement]
for /f "tokens=1" %%m in ('%uf% header list 72850817000000000000000000000002') do (
<nul set /p TmpStr=mCode FFS:
%ur% %%m 1 %modcpu% -o bios.bin -asis || move tmp\bios.bak bios.bin>nul && echo File restored && pause && goto cpu
rem %mmt% /r %%m %modcpu%
)
By
2
3
4
5
6
echo [Replacement]
for /f "tokens=1" %%m in ('%uf% header list 72850817000000000000000000000002') do (
<nul set /p TmpStr=mCode FFS:
rem %ur% %%m 1 %modcpu% -o bios.bin -asis || move tmp\bios.bak bios.bin>nul && echo File restored && pause && goto cpu
%mmt% /r %%m %modcpu%
)
And I used option M
@tistou77
Thank you.
I checked the differences in the assemblies and the problem in TE Section. MMT processes them in its own way. We’ll keep thinking.
In the last b6, I made the choice of a tool for updating.
Latest 306A9 microcode for IVB is not being detected as latest nor found in db
After option C to Create FFS
Note: Microcode #1 was not found at the database, please report it!
Edit: FYI I tested updating the mCode using both UEFIReplace and MMTool and the binaries are identical using hash/compare tools
There is one idea, I’ll make a file later.