"Mo replacements…"? It’s not a problem. Look at the version of the files. You have 0.0.19 and to update 0.0.19. If the files are the same, then UEFIReplace will not update anything.
HI EVERYONE… Sorry for my English… Tested today Update v1.70 b5 to modify Asus Z97 Pro Gamer BIOS v.2203 with latest CPU microcode, which is version 25 for Hasswell CPUs… after flash modifyed BIOS with b5, my PC dont turn on…i have extra bios chip with factory bios, so i reflashed bad chip by replacing it after i start my pc with good one… tried to update cpu microcode only and nothing else, same thing - PC NOT BOOTING at all, all fans at max speed and black screen… updated BIOS with UEFI BIOS Updater Update v1.70 b2 Dev to CPU Microcode version 24 , flashed bios chip and everything ok, PC booted with no any problem. Thank you EVERYBODY, who keep UEFI BIOS Updater alive and work hard, and please check latest Update b5, probably someone else will face boot problem after flashing their bios, modified with beta 5 and updated Hasswell CPU Microcode to version 25…Thank you
We need to find out whether such a problem (code 55) when using MMTool. This will change the utility from UEFIReplfce to MMTool It is desirable that the MMTool version is 5.2.0.24.
But then it is very strange that no one wrote about such problems using 1.70b2. The difference between b2 and b5, when updating microcodes, no.
Add: Probably I will update the menu, adding a choice:
1 2 3 4 5
u1 - Update with UEFIReplace u2 - Update with MMTool 0 - Cancel
I assume that you have a Haswell 306C3 CPU? The version 25 microcode is Intel’s v2 for fixing the Meltdown and Spectre vulnerabilities. Intel released v2 in April 2018. The version 24 microcode is Intel’s v1 for fixing the Meltdown and Spectre vulnerabilities. Intel released v1 in November 2017??? Note that v1 is quite buggy. The v1 microcode causes CPUs to generate tons of recoverable errors. The number of errors goes up if you slightly overclock. Even though these errors were recoverable, they can result in data corruption. The v1 microcode occasionally generates an unrecoverable CPU error which causes the computer to reboot. I need to test exactly which is the latest microcode for the Haswell 306C3 which did not try to fix Meltdown and Spectre.
Best regards, –GTP
EDIT by Fernando: Directly addressing prefix added (to make clear, to whom the author has answered)
I have moved your posts into this better matching “UBU Discussion” thread. The other UBU thread is reserved for the Guide and Sonix’s announcements of new/updated UBU versions. Thanks for your understanding!
By tests, this is the first and buggy microcode that protects against these vulnerabilities. The last microcode is not contained in itself protection against vulnerabilities it is 0x22.
Guys i can confirm that CPU Micro-Code Rev. 25 for Haswell CPUs is not working correctly and when integrated into BIOSes the machine can’t boot, so STAY AWAY FROM REV. 25 CPU MICRO-CODE FOR HASWELL!!
This is a riot of fun, since the April 2018 revision 25 for Haswell CPUs is supposed to be Intel’s grand and final fix for Meltdown and Spectre for their Haswell CPUs! I say this since Intel now claims that they basically have ended all work with regards to Meltdown and Spectre. I see the lawsuits a coming to fruition. It is as if Intel has taken a page out of Microsoft Nadella’s book in order to turn the general public into beta testers.
Hmm…sounds like the UEFI BIOS Updater should be released into two specific versions:
A pre-Spectre version, with the latest pre-Spectre Intel microcode updates.
A post-Spectre version, with the latest and supposedly working Intel microcode patches for Meltdown and Spectre. Obviously this post-Spectre version of the UEFI BIOS Updater should be for the faint of heart.
Lest you all forget, there is yet another Spectre related vulnerability, called BranchScope, which Intel has yet to address.
Note these two things:
Microsoft, via Windows updates, is protecting all Windows OS versions from Meltdown.
Presently there are no known “in the wild” exploits of Spectre. My caveat: Just because no Spectre expoits have been detected, this does not mean that they do not exist, and this does not mean that Spectre exploits may be embedded inside newly downloaded programs. The upshot is to be very careful of what programs you choose to download and install.
Overall, isn’t this a riot of fun?
Microsoft has tried to step up to the plate by baking Intel’s microcode for preventing both Meltdown and Spectre into Windows 10. Yet Microsoft has chosen to NOT bake Spectre protection into Windows 7 or Windows 8.1. Why? Isn’t this obvious? Because Microsoft wants everyone to use Windows 10 which is their grand Telemetry Operating System which, out of the box, both records and reports everything which you do with your computer. So far, Microsoft has only baked in Meltdown (and no Spectre) protection for Windows 7 and Windows 8.1 via the latest Windows updates.
Finally, I am trying to say the following:
Given the present situation that Intel has obviously not thoroughly tested its microcode updates which are meant to prevent Meltdown and Spectre CPU vulnerabilities, everyone should be very careful about applying any late 2017 and 2018 BIOS microcode updates.
@Maison1 : Glad to hear that for you all went smoothly but in my case i’ve used UEFI BIOS Updater Update V.1.70 B5 Dev and got no POST after BIOS flash…
@SoniX : OK SoniX and thanks a lot for your hard work.
Haswell CPU Micro-Code Rev. 25 brings hardware support for CVE-2018-3639
Speculation control settings for CVE-2018-3639 [speculative store bypass]
Hardware is vulnerable to speculative store bypass: True Hardware support for speculative store bypass mitigation is present: True Windows OS support for speculative store bypass mitigation is present: True Windows OS support for speculative store bypass mitigation is enabled system-wide: False