TE
TechEcho
Home24h TopNewestBestAskShowJobs
GitHubTwitter
Home

TechEcho

A tech news platform built with Next.js, providing global tech news and discussions.

GitHubTwitter

Home

HomeNewestBestAskShowJobs

Resources

HackerNews APIOriginal HackerNewsNext.js

© 2025 TechEcho. All rights reserved.

The AMD Branch (Mis)Predictor Part 2: Where No CPU Has Gone Before

152 pointsby gbrown_about 3 years ago

11 comments

cwilluabout 3 years ago
The final pithy quote about how a prominent linux developer claimed no cpu would ever do such a thing, loses some of its impact when you realize that a _more_ prominent linux developer replied saying “I think, when it comes to speculative execution, our general expectation that CPUs don't do idiotic things got somewhat weakened in the past year or so ...”
评论 #30611467 未加载
评论 #30612708 未加载
CorrectHorseBatabout 3 years ago
Too bad Grsecurity isn't able to work together with upstream, their security knowledge seems to be legit.
评论 #30606590 未加载
评论 #30612790 未加载
userbinatorabout 3 years ago
To my knowledge none of these side-channels, even the original ones several years ago, have been exploited practically in the wild. IMHO gathering the amount of detail needed to attempt such an attack, as exemplified by the demos that have been given, would itself be prohibitively difficult. Thus the impact of yet another one remains negligible to a personal computer user, but of course the cloud providers would be super-paranoid about it.<p>It&#x27;s worth noting that the memory protection scheme, introduced with the 286, was never intended to be a strong security barrier, but instead a means of isolating bugs and making them easier to debug.
评论 #30610372 未加载
Lascailleabout 3 years ago
That&#x27;s a great article.<p>I do worry sometimes that something is up with CPU development, that we&#x27;re tending towards more and more complicated designs with workflows that are very hard to analyse and simulate even for the designers themselves, but the actual workload execution ability performance per core isn&#x27;t shifting upwards all that much, and then weird mitigations have to be applied that reduce that execution ability in practice.<p>Something makes me think that perhaps a different design paradigm should prevail, with particular attention paid to segregation of workloads and of core partitioning, perhaps an abandoning of hyperthreading and even to the extent of having 100% physical separation of cores and their caches.<p>But I&#x27;m very much not an expert in the field.<p>A little birdie inside of me every now and then wakes up and whispers &#x27;is it a coincidence that these design paradigms are yielding so many vulnerabilities?&#x27;
评论 #30611233 未加载
评论 #30611254 未加载
评论 #30609864 未加载
throwawaylinuxabout 3 years ago
AMD is supposed (or does) have very good branch predictors. What&#x27;s interesting to me is they don&#x27;t do a re-steering of the wrong prediction well before later instructions are issued to the back end.<p>Maybe their BTB is that good that they didn&#x27;t see it worth investing in the control logic for it.
RcouF1uZ4gsCabout 3 years ago
SPS Theorem<p>Pick two:<p>Shared Physical Hardware<p>Performance<p>Security
评论 #30605806 未加载
评论 #30664290 未加载
B1FF_PSUVMabout 3 years ago
Somehow the &quot;Final remarks&quot; section reminded me of the more paranoid &quot;wheels within wheels&quot; Frank Herbert fiction.<p>Not pleasantly, the FH fiction wasn&#x27;t, either.
unixbaneabout 3 years ago
the funny part is that PCs are shown to be unusuable for security sensitive things as usual and all this research will be used to optimize video games
morpheuskafkaabout 3 years ago
What does &quot;code gadget&quot; mean in this context?
评论 #30612755 未加载
评论 #30610981 未加载
divyekapoorabout 3 years ago
Very well done.
jimmaswellabout 3 years ago
I went AMD for the first time with my new laptop. Some 8 core Ryzen. Games crash all the time, not sure if it&#x27;d the architecture&#x27;s fault, and I can&#x27;t properly virtualize Windows 98 (which I like to just for fun&#x2F;nostalgia) apparently due to the architecture. I feel like I&#x27;m sticking with Intel from now on like my old gut instinct said to.
评论 #30610882 未加载