Discussion in 'Article Discussion' started by bit-tech, 3 Jan 2018.
It might finally be time to switch to AMD.
Depending on workload, the performance impact can be absolutely nothing. Unless your workload is absolutely hammering syscalls, then you're probably not going to be affected to a measurable degree.
A big headache for servers, but for desktop use unlikely to cause any problems.
That certainly appears to be the case for Linux but it maybe a little hasty to extrapolate that to Windows as it's my understand the way Linux and Windows deal with drivers maybe markedly different, in that Linux runs drivers in user space so something like games that depend on drivers doesn't do much, if any, context switching whereas the picture for Windows drivers is less clear.
I know nothing about driver development for Windows though so maybe the context switching is minimal.
Microsoft's documentation on the subject says the following...
Hence "can result" - i.e. "its implementation can result in a performance loss and where such a performance loss occurs it is between five and 30 percent depending on workload." Sorry if I didn't phrase that bit very clearly!
Windows has had user-mode drivers since Windows Vista a decade ago. The big switch was what people threw a fit about when manufacturers declined to update their drivers to maintain compatibility.
It's had partial user-mode drivers, there's still large portions of drivers handled in Kernel-mode.
AMD is not effected but will suffer the same performance hit due to the patch. its been confirmed now
i would be interested for bit-tech to run benchmarks on their test rig pre and post patch, just to see if any games are actually affected
...not just Intel, some ARM cpus are affected too [apparently]
"To be clear, the security research team identified three variants targeting speculative execution," AMD said in a statement to Axios. "The threat and the response to the three variants differ by microprocessor company, and AMD is not susceptible to all three variants. Due to differences in AMD's architecture, we believe there is a near zero risk to AMD processors at this time. We expect the security research to be published later today and will provide further updates at that time."
don't see how that article suggests AMD are affected too, guess since the vulnerability hasn't been released for obvious reasons that may yet change.
It is worth including the sentence from the article before your quote:
Also i did say ARM, not AMD
There's actually TWO vulnerabilities.
One "only" hits every Intel part since Pentium Pro(Itanium and pre-2013 Atoms excepted). The other hits AMD, Intel, and ARM.
Both have supervillain names.
How lucky it was for the Inel CEO to sell all those shares he had to take him down to the 250,000 bare minimum he needs under his employment agreement before all this broke.
Google have posted details on their project zero blog.
AMD is not susceptible to Meltdown. ARM, AMD, Intel is to Spectre. Likely custom versions of ARM cores too.
Official statement from AMD on how each variant effects them.
Variant One Bounds Check Bypass Resolved by software / OS updates to be made available by system vendors and manufacturers. Negligible performance impact expected.
Variant Two Branch Target Injection Differences in AMD architecture mean there is a near zero risk of exploitation of this variant. Vulnerability to Variant 2 has not been demonstrated on AMD processors to date.
Variant Three Rogue Data Cache Load Zero AMD vulnerability due to AMD architecture differences.
Official statement from ARM listing what models are effected.
No statement from Intel about what variants effect what CPUs as of yet.
So what can we do, dust off those old Pentiums and K6s? I think 32-bit linux and Win98 should still be running well on them.
Separate names with a comma.