181
DXGI debugging: Microsoft put me on a list
Why does Space Station 14 crash with ANGLE on ARM64? 6 hours later…
New
The latest blog posts to take off on Hacker News. 543 posts.
182
Disk Utility still can't check and repair APFS volumes and containers (2021)
Don’t waste time trying to discover which files are still open when you want to check or repair an APFS disk and Disk Utility throws an error. Here’s the solution.
183
Systemd can be a cause of restrictions on daemons
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…
184
Processing Strings 109x Faster Than Nvidia on H100
StringZilla v4 brings CUDA acceleration for string processing: 109x faster than Nvidia's CuDF on edit distances, plus 52-bit MinHash fingerprinting and AES-based hashing.
185
TernFS – An exabyte scale, multi-region distributed filesystem
XTX Markets is a leading algorithmic trading firm which uses state-of-the-art machine learning technology to produce price forecasts for over 50,000 financial instruments across equities, fixed income, currencies, commodities and crypto. It uses those forecasts to trade on exchanges and alternative trading venues, and…
186
I uncovered an ACPI bug in my Dell Inspiron 5567. It was plaguing me for 8 years
Imagine you close your laptop lid to put it to sleep, but instead of pausing, it reboots. Not every time, just often enough to be infuriating. You try to save your work, but the machine decides to start over.
187
Hypervisor 101 in Rust
This is a day long course to quickly learn the inner working of hypervisors and techniques to write them for high-performance fuzzing.
188
A postmortem of three recent issues
A technical report on three bugs that intermittently degraded Claude's responses: what happened, why fixes took time, and what we're changing.
189
Optimizing ClickHouse for Intel's ultra-high core count processors
Intel's latest processor generations are pushing the number of cores in a server to unprecedented levels. For analytical databases like ClickHouse, ultra-high core counts represent a huge opportunity and a complex challenge.
190
Stategraph: Terraform state as a distributed systems problem
Terraform state is a distributed systems problem masquerading as file storage. Graph state fixes the bottlenecks.
191
Implicit ODE solvers are not universally more robust than explicit ODE solvers
A very common adage in ODE solvers is that if you run into trouble with an explicit method, usually some explicit Runge-Kutta method like RK4, then you should try an implicit method. Implicit methods, because they are doing more work, solving an implicit system via a Newton method having “better” stability, should be…
192
Writing an operating system kernel from scratch
Exploring a minimal implementation of a time-sharing kernel on RISC-V, implemented in Zig, on top of OpenSBI.
193
RIP pthread_cancel
I posted about adding pthread_cancel use in curl about three weeks ago, we released this in curl 8.16.0 and it blew up right in our faces. Now, with #18540 we are ripping it out again. What happened? short recap pthreads define “Cancelation points”, a list of POSIX functions where a pthread may be cancelled. In…
194
195
Tips for installing Windows 98 in QEMU/UTM
Windows 98 runs surprisingly well in QEMU via UTM SE, but it requires some care in setting it up. It’s a great way to run old 90s Windows and DOS software on your iPad (and Mac too, though you have… Continue reading →
196
Removing newlines in FASTA file increases ZSTD compression ratio by 10x
First released with Zstandard 1.3.2 in 2017, the --long range match finder increases the compressor’s search window to at least 128MiB, improving deduplication inside large files. This optional feature had substantial performance overheads at launch, but various optimisations have since brought its performance within…
197
PythonBPF – Writing eBPF Programs in Pure Python
I brag about my newest obsession.
198
199
NT OS Kernel Information Disclosure Vulnerability
Bidding farewell to one of the last kernel address leaks, CVE-2025-53136. Even patches can open new doors for exploitation.
200
Performance Improvements in .NET 10
Take a tour through hundreds of performance improvements in .NET 10.
201
Hypervisor in 1k Lines
Write your first hypervisor from scratch, in 1K LoC.
202
How Ruby executes JIT code
Where does JIT-compiled code live? How does Ruby switch between bytecode and native execution? Why does TracePoint slow everything down? This post answers the JIT questions most Ruby developers have but rarely see explained.
203
Memory Integrity Enforcement
Memory Integrity Enforcement (MIE) is the culmination of an unprecedented design and engineering effort spanning half a decade that combines the unique strengths of Apple silicon hardware with our advanced operating system security to provide industry-first, always-on memory safety protection across our devices …
204
I solved a distributed queue problem after 15 years
Learn how queues make horizontal scaling, scheduling, and flow control easier in cloud systems, and how to make them durable and observable.
205
Shipping textures as PNGs is suboptimal
Are you shipping textures to players as PNGs? The goal of this post is to convince you that this is suboptimal, and walk you through a better approach.
206
Default musl allocator considered harmful to performance
In a real world benchmark, the default musl allocator caused a 7x slowdown compared to other allocators. I recommend all Rust projects immediately swap to a different allocator in a musl environment.
207
Firefox 32-bit Linux Support to End in 2026
For many years, Mozilla has continued to provide Firefox for 32-bit Linux systems long after most other browsers and operating systems ended support. We made this choice because we care ...
208
Fil's Unbelievable Garbage Collector
Fil-C uses a parallel concurrent on-the-fly grey-stack Dijkstra accurate non-moving garbage collector called FUGC (Fil's Unbelievable Garbage Collector). You can find the source code for the collector itself in fugc.c , though be warned, that code cannot possibly work without lots of support logic in the rest of the…
209
io_uring is faster than mmap
Sourcing data directly from disk IS faster than caching in memory. I brought receipts. Because hardware got wider but not faster, the old methods don't get you there. You need new tools to use what is scaling and avoid what isn't.
210
Hitting Peak File IO Performance with Zig
This post goes through how to maximize file IO performance on linux using zig with io_uring.