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.

CPU utilization is wrong [video]

45 pointsby startupflixabout 7 years ago

5 comments

DonbunEf7about 7 years ago
This nigh-useless blurb links to a Youtube video: <a href="https:&#x2F;&#x2F;youtu.be&#x2F;QkcBASKLyeU" rel="nofollow">https:&#x2F;&#x2F;youtu.be&#x2F;QkcBASKLyeU</a>
评论 #16960650 未加载
gnodeabout 7 years ago
To save you the time of watching the video: Spectre &#x2F; Meltdown mitigations (page table isolation) cause execution inefficiency because of TLB cache misses.
dragontamerabout 7 years ago
So first of all: the lightning talk is certainly &quot;worth it&quot;. Brendan Gregg is very knowledgeable and I can certainly recommend watching the talk. Its a good overview into the mindset of low-level optimization.<p>However, the core premise, the &quot;headline&quot; if you will of this talk, is utterly wrong. I get it: the headline needs to be a bit provocative to draw eyes, but its closer to &quot;clickbait&quot; than a good title of the talk.<p>The TL;DR of the talk is: Spectre&#x2F;Meltdown caused a slowdown. But said slowdown is coming up as a ~+20% &quot;CPU Utilization&quot; under Linux performance measurement (ie: top or ps). This slowdown is because of a severe increase in TLB-misses.<p>Which... doesn&#x27;t seem &quot;wrong&quot; to me. If the CPU is stalling because of TLB-buffer misses (kinda similar to an L2 cache miss or whatnot), then that&#x27;s certainly CPU-time that is being wasted on something. Perhaps it isn&#x27;t using power since the CPU is fully stalled, but the CPU is certainly &quot;utilized&quot; during that timeframe. Aside from hyperthreading: you can&#x27;t really do something during the stall.<p>----------<p>Anyway, my criticism aside, the talk is certainly worthwhile to watch. And that&#x27;s the important bit. So go ahead and watch the talk, even if I have a few pedantic qualms about the phrasing of the headline.<p>The real benefit of the talk is how Brendan Gregg diagnosed this performance problem, which eventually led to his conclusion about TLB-misses + Spectre. Demonstrating this skill and walking through a system to be able to draw this kind of conclusion is incredibly important.<p>I just don&#x27;t... really like the title of the talk I guess. A rather minor concern all else considered.
评论 #16960912 未加载
billysieluabout 7 years ago
So is the CPU available to do something else or not?
评论 #16961748 未加载
gfnordabout 7 years ago
That&#x27;s great and I agree with him. We have to pay attention to the stale.