Free Threaded Python with Asyncio(blog.changs.co.uk)

6 points by rbanffy 2 years ago | 0 comments

%CPU utilization is a lie(brendanlong.com)

396 points by BrendanLong 1 year ago | 135 comments

Blog post by Brendan Long arguing that %CPU utilization reported by tools like top misrepresents how close servers are to their actual capacity limits.

Tokio and Prctl = Nasty Bug(kobzol.github.io)

190 points by ingve 1 year ago | 61 comments

CPU Utilization Is Wrong (2017)(brendangregg.com)

75 points by xk3 3 years ago | 38 comments

Jacobin: A more than minimal JVM written in Go(jacobin.org)

268 points by yla92 3 years ago | 162 comments

How to optimize a CUDA matmul kernel for cuBLAS-like performance (2022)(siboehm.com)

86 points by mpweiher 2 years ago | 21 comments

Why is my CPU usage always 100%?(downtowndougbrown.com)

505 points by pncnmnp 1 year ago | 137 comments

Part 9 of a blog series on upgrading the Chumby 8's Linux kernel from 2.6.28 to 6.x, investigating why top always reports 100% CPU usage.

Branch Privilege Injection: Exploiting branch predictor race conditions(comsec.ethz.ch)

397 points by alberto-m 1 year ago | 177 comments

ETH Zurich COMSEC research page on Branch Privilege Injection, a class of Intel CPU vulnerabilities exploiting branch predictor race conditions to leak privileged memory.

C++26: The New Debugging Header(sandordargo.com)

98 points by jandeboevrie 9 days ago | 23 comments

Blog post by Sandor Dargo introducing the C++26 <debugging> header for programmatic breakpoints, replacing compiler-specific code like __debugbreak.

Zenbleed(lock.cmpxchg8b.com)

1160 points by loeg 3 years ago | 327 comments

Tavis Ormandy's writeup of Zenbleed, an AMD Zen 2 CPU vulnerability that leaks data from vector registers, covering exploitation, discovery, and mitigation.

The Lost Atomic Update on Loongson CPU(jia.je)

120 points by jiegec 1 month ago | 6 comments

A blog post investigating a hardware erratum on the Loongson LA664 CPU where an atomic instruction loses updates, causing an infinite loop while packaging normaliz for Debian.

How fast are Linux pipes anyway? (2022)(mazzo.li)

462 points by SEJeff 3 years ago | 87 comments

Blog post exploring how Linux pipes are implemented by iteratively optimizing a pipe-based test program from 3.5GiB/s to about 65GiB/s using perf profiling.