• Competitor rules

    Please remember that any mention of competitors, hinting at competitors or offering to provide details of competitors will result in an account suspension. The full rules can be found under the 'Terms and Rules' link in the bottom right corner of your screen. Just don't mention competitors in any way, shape or form and you'll be OK.

Intel bug incoming? Meltdown and Spectre exploits

I'm not 100% sure on this. But reading a few other sites it appears that Linux can load CPU microcode updates at startup. So if you're running Linux then you may not need a BIOS update to mitigate Spectre.
 
I'm not 100% sure on this. But reading a few other sites it appears that Linux can load CPU microcode updates at startup. So if you're running Linux then you may not need a BIOS update to mitigate Spectre.
This is correct from my own testing.
 
I'm not 100% sure on this. But reading a few other sites it appears that Linux can load CPU microcode updates at startup. So if you're running Linux then you may not need a BIOS update to mitigate Spectre.
Windows can do something similar but you still need the microcode so not sure where you get that from?
 
I ran some benchmarks this evening, before and after applying KB4056892 on Windows 10, 1709 with a 6700K.

comp.PNG


Sorry my freebie security certificate has expired, will fix tomorrow... its only a .png.

Not much difference across the board, apart from the Crystal Disk Mark results, -11% in some situations.
Realbench fell 1.8%, Cinebench fell about 1.5%, Superposition gained 0.14%.

Basically all fairly within expected benchmark deviations, excluding the Crystal Disk Mark results.
 
I'm not 100% sure on this. But reading a few other sites it appears that Linux can load CPU microcode updates at startup. So if you're running Linux then you may not need a BIOS update to mitigate Spectre.

Yes, microcode updates are usually delivered via Linux package managers, and you can verify from logs (e.g. if you're on Ubuntu run dmesg) that the kernel applies updates as part of booting:

$ dmesg | grep microcode
[ 1.095824] microcode: CPU0: patch_level=0x08001126
[ 1.095844] microcode: CPU1: patch_level=0x08001126
[ 1.095865] microcode: CPU2: patch_level=0x08001126
[ 1.095873] microcode: CPU3: patch_level=0x08001126
[ 1.095888] microcode: CPU4: patch_level=0x08001126
[ 1.095899] microcode: CPU5: patch_level=0x08001126
[ 1.095916] microcode: CPU6: patch_level=0x08001126
[ 1.095926] microcode: CPU7: patch_level=0x08001126
[ 1.095943] microcode: CPU8: patch_level=0x08001126
[ 1.095951] microcode: CPU9: patch_level=0x08001126
[ 1.095956] microcode: CPU10: patch_level=0x08001126
[ 1.095959] microcode: CPU11: patch_level=0x08001126
[ 1.095975] microcode: CPU12: patch_level=0x08001126
[ 1.095985] microcode: CPU13: patch_level=0x08001126
[ 1.095995] microcode: CPU14: patch_level=0x08001126
[ 1.096003] microcode: CPU15: patch_level=0x08001126
[ 1.096031] microcode: Microcode Update Driver: v2.2.
 
From everything I've read computers will need a BIOS upgrade to counter Spectre type attacks? If so is there any news if Gigabyte will be releasing BIOS updates?
 
I was sorting out a new build around the 8700k but I really don't know what to do now.

If you're not intending to run performance-sensitive database workloads or VMs you shouldn't be worried about the performance impact of the Meltdown patch.

I find it really hard to believe this level of flaw went unnoticed for so long.

Side channel attacks are a relatively recent avenue of research. Remember that up to fairly recently there were so many other simpler and easier to exploit vulnerabilities out there such as buffer overflows etc. to keep everyone busy, it was more important to tackle those first. Perhaps instead it's a sign that as the average level of security hardening increases, research moves onto more obscure avenues of attack; of course this'll inevitably throw up new classes of vulnerability that were not widely known (or at least proven) over time.
 
I find it really hard to believe this level of flaw went unnoticed for so long.

On the desktop end it's been there since at least the Pentium 4 x64, quite amazing this has been left like this so long. Think of all the old computers in industry that are running useful software but not really supported any more, I expect vast numbers of computers will never even get any security patches.
 
From everything I've read computers will need a BIOS upgrade to counter Spectre type attacks? If so is there any news if Gigabyte will be releasing BIOS updates?

Doubt there will be much of substance from the AIBs, etc. until after the 9th when they will get full information - even then I'll be surprised if there is much in the way of timely updates, etc.
 
Not sure if this is of interest, but just checked Nvidia Quadro drivers, and the latest Quadro drivers have a fix for the speculative execution issue in them, not sure about Geforce drivers.

http://www.nvidia.co.uk/download/driverResults.aspx/129031/en-uk

even then I'll be surprised if there is much in the way of timely updates, etc.

Yes I expect it will take ages to get BIOS updates rolled out, I expect some older motherboards may never get updates.
 
Back
Top Bottom