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.

KUtrace: Low-overhead Linux kernel tracing facility

172 pointsby luu10 months ago

10 comments

ndesaulniers10 months ago
Early in my career, I reached out to the author and was able to grab lunch with him; he was about to retire! Was insightful to hear his thoughts on system performance, particularly systems involving more than one machine, which is something they studied deeply.<p>It gave me appreciation for the amount of knowledge one accumulates over a career, and what a loss it is to an organization when one so knowledgeable retires.
crissaegrim10 months ago
I had the pleasure to work with Dick on getting KUtrace to work on Android devices last year. It was a great experience to work with one of the greats in systems performance. He was a wealth of information regarding performance bottlenecks and optimizations.<p>KUtrace is absolutely one of the most powerful tools I&#x27;ve used for deeply understanding performance bottlenecks (after isolating issues) such as poor scheduling behavior. I would highly recommend reading his book &quot;Understanding Software Dynamics&quot; [1] if you are interested in learning more about KUtrace or performance bottlenecks&#x2F;optimizations in general. The book is quite dense and dives deep into the performance characteristics of many examples of the five fundamental resources (according to Dick): CPU, Memory, Disk&#x2F;SSD, Network, and Software critical sections.<p>[1]: <a href="https:&#x2F;&#x2F;www.oreilly.com&#x2F;library&#x2F;view&#x2F;understanding-software-dynamics&#x2F;9780137589692&#x2F;" rel="nofollow">https:&#x2F;&#x2F;www.oreilly.com&#x2F;library&#x2F;view&#x2F;understanding-software-...</a>
评论 #40981490 未加载
评论 #40976829 未加载
评论 #40983314 未加载
评论 #40974619 未加载
CalChris10 months ago
Site&#x27;s article <i>Benchmarking &quot;Hello, World!&quot;</i> is basically a KUTrace tutorial.<p><a href="https:&#x2F;&#x2F;queue.acm.org&#x2F;detail.cfm?id=3291278" rel="nofollow">https:&#x2F;&#x2F;queue.acm.org&#x2F;detail.cfm?id=3291278</a><p>Also, <a href="https:&#x2F;&#x2F;www.youtube.com&#x2F;watch?v=D_qRuKO9qzM" rel="nofollow">https:&#x2F;&#x2F;www.youtube.com&#x2F;watch?v=D_qRuKO9qzM</a>
li4ick10 months ago
His book, &quot;Understanding Software Dynamics&quot;, is one of the best technical books I&#x27;ve ever read. Top 3 for me.
评论 #40974820 未加载
usr110610 months ago
Sounds very interesting.<p>But it works by patching the kernel, not just using eBPF like many performance tools recently. So it needs active maintenance all the time considering the current velocity of internal kernel changes. And I would not be surprised if it didn&#x27;t build or work correctly if you have a heavily patched and customized kernel.<p>On the positive side at a first glimpse the maintenance to adapt to new kernels looks very active.
c0deR3D10 months ago
Out of curious, does BPF now capable of capturing all the context switch events such as CPU trap?<p>Also, if the overhead is negligible, maybe the author can try to merge this into mainline with the use of static key to make the incurred overhead switchable. In spite of the static key, the degree of the accompanied inteferences on cache and branch predictor might be an intriguing topic though.
评论 #40972861 未加载
mgaunard10 months ago
Why not just use the eBPF system?
评论 #40977224 未加载
mydriasis10 months ago
I thought this read &quot;training facility&quot;, and I was excited to sign up!
xoranth10 months ago
How would this interact with `io_uring`, especially the polling modes (IO_SETUP_SQPOLL, IO_SETUP_IOPOLL)?
评论 #40974515 未加载
elric10 months ago
Is the naming intentional? Or just a weird coincidence? Kut being Dutch for cunt, by an author called dick..?
评论 #40974262 未加载
评论 #40980805 未加载