121
Mo RAM, Mo Problems (2025)
As a retro-computer enthusiast, it seems that parts are either insanely expensive or dirt cheap. If the first case has obvious problems, the second can also lead to issues.
122
Back to FreeBSD: Part 1
Containers ran in production on FreeBSD thirteen years before Docker. So why does nobody talk about it?
123
Molly guard in reverse
A blog about software craft and quality
124
What makes Intel Optane stand out (2023)
Once in a while new hardware is released that makes a difference. Such a device is the Intel Optane series of high-performance SSD’s for professional use, which was released in late 2017. In this case I’m talking about the Intel Optane P4800X and P5800X and their consumer counterparts ( 900P and 905P ). All drives are…
125
Running out of disk space in production
A tale about misconfiguring nginx, from adventures in production
126
Backpressure is all you need
Hot takes and cold truths on software, startups, and the lies we tell ourselves.
127
Raspberry Pi Drag Race: Pi 1 to Pi 5 – Performance Comparison
Today we're comparing one of each generation of Pi from the Pi 1 to the Pi 5 to see how their performance and specifications have improved.
128
Five years of running a systems reading group at Microsoft
I started a reading group in 2021, a few months after joining Microsoft as a new grad on the Azure Databases team. The group was initially focused on database internals, which was my favorite subject at UW. Databases touch so many areas of CS: compiler construction in the query engine, memory management with the…
129
Getting 50 GB/S Back from the Apple Neural Engine
An RTL performance erratum in the Apple M3 Neural Engine throttles DRAM weight streaming throughput down to 17–19 GB/s from the nominal 45–60 GB/s, whenever the total weight size is an integer multiple of 1 MiB, which currently affects 7 of ANEMLL ’s 15 models. Avoiding the problematic path in the kernel DMA engine's…
130
Branchless Quicksort faster than std:sort and pdqsort with C and C++ API
Performance results naturally depend on the underlying hardware. The following benchmarks show the execution times for sorting 50 million doubles using different sorting implementations. The measurements were taken on an Apple M1 system using Clang and on an AMD Ryzen 3 Linux system using GCC , both compiled with the…
131
Azure Linux 4.0 is Microsoft's first general-purpose Linux
Microsoft’s in-house Linux, the distribution that grew out of CBL-Mariner, just hit public preview as a general-purpose cloud OS you can run on any Azure VM and coming soon to WSL.
132
133
Debunking Zswap and Zram Myths
zswap and zram are fundamentally different approaches with different philosophies. If in doubt, use zswap.
134
135
Switch Lowering in GCC
We have seen in a previous post that switch statements are more complex than they seem. The compiler is able to implement them as a decision tree (even though maybe it shouldn't). Is that everything it can do?
136
C++26 Shipped a SIMD Library Nobody Asked For
std::simd Can't Express 90% of Real SIMD Code
137
Caddy compatibility for zeroserve: 3x throughput and 70% lower latency
zeroserve is a high-performance HTTPS server that runs eBPF scripts in userspace ( intro ). Now it's got a Caddy-compat mode - when provided a Caddyfile, zeroserve JIT-compiles it to eBPF and then to native x86_64/ARM64 machine code, and runs it in an io_uring event loop.
138
How Compaction Works in Pi
Why compaction is needed for large language models and how Pi implements it
139
The Ubuntu 26.04 Rust sudo is not compatible with standard sudo
You're probably reading this page because you've attempted to access some part of my blog (Wandering Thoughts) or CSpace , the wiki thing it's part of. Unfortunately you're using a browser version that my anti-crawler precautions consider suspicious, most often because it's too old (most often this applies to versions…
140
A simplified model of Fil-C
I've seen lots of chatter about Fil-C recently, which pitches itself as a memory safe implementation of C/C++. You can read the gritty details of how this is achieved, but for people coming across it for the first time, I think there is value in showing a simplified version, as once you've understood the simplified…
141
Diskless Linux boot using ZFS, iSCSI and PXE
Diskless Linux Boot
142
My Snapdragon Dev Kit was healthy and working fine until a Windows update failed
My personal website and blog
143
Running Your Own AS: BGP on FreeBSD with FRR, GRE Tunnels, and Policy Routing
How I obtained my own AS number and IPv6 prefix, set up a FreeBSD BGP router with FRR, and built a tunnel overlay to bring globally routable addresses to servers that already have provider-assigned...
144
Run OpenBSD on DigitalOcean for $4/month
My homepage now runs on OpenBSD with httpd(8) and Let's Encrypt via acme-client(1) . I previously hosted it on GitHub Pages and then eventually moved to Cloudflare Pages because both options were free and easy. Free and easy is cool, and I understand that writing software full-time leaves us wanting absolutely nothing…
145
How to make Firefox builds 17% faster
In the previous post, I mentioned that buildcache has some unique properties compared to ccache and sccache. One of them is its Lua plugin system, which lets you write custom wrappers for programs that aren’t compilers in the traditional sense. With Bug 2027655 now merged, we can use this to cache Firefox’s WebIDL…
146
Girls just wanna have fast MPMC queues with bounded waiting
Disclaimer: An earlier version of this post claimed the structure is wait-free, this is incorrect. Being wait-free requires that failure or suspension of any thread can’t cause failure or suspension of another thread. This queue in fact does not fulfill that requirement. The main section which discusses the wait…
147
15 years of FP64 segmentation, and why the Blackwell Ultra breaks the pattern
Buy an RTX 5090, the fastest consumer GPU money can buy, and you get 104.8 TFLOPS of FP32 compute. Ask it to do double-precision math and you get 1.64 TFLOPS. That 64:1 gap is not a technology limitation. For fifteen years, the FP64:FP32 ratio has been slowly getting wider on consumer GPUs, widening the divide between…
148
149
Unikernels were hard. key word: were
Justin Cormack, who worked on MirageOS back in the day, called me at the pub to talk about why unikernels are coming back. The friction that made them hard is gone, the operating system is design debt, and if you care about security you should seriously consider them.
150
Crawling a billion web pages in just over 24 hours, in 2025
For some reason, nobody's written about what it takes to crawl a big chunk of the web in a while: the last point of reference I saw was Michael Nielsen's post from 2012 .