The Illustrated Transformer
jalammar.github.io
Ultrasound Cancer Treatment: Sound Waves Fight Tumors
spectrum.ieee.org
GLM-4.7: Advancing the Coding Capability
z.ai
Flock Exposed Its AI-Powered Cameras to the Internet. We Tracked Ourselves
404media.co
The Garbage Collection Handbook
gchandbook.org
NIST was 5 μs off UTC after last week's power cut
jeffgeerling.com
Claude Code gets native LSP support
github.com
Scaling LLMs to Larger Codebases
blog.kierangill.xyz
Lotusbail npm package found to be harvesting WhatsApp messages and contacts
koi.ai
Show HN: C-compiler to compile TCC for live-bootstrap
github.com
Universal Reasoning Model (53.8% pass 1 ARC1 and 16.0% ARC 2)
arxiv.org
How the RESISTORS put computing into 1960s counter-culture
spectrum.ieee.org
Satellites reveal heat leaking from largest US cryptocurrency mining center
space.com
The biggest CRT ever made: Sony's PVM-4300
dfarq.homeip.net
US blocks all offshore wind construction, says reason is classified
arstechnica.com
There Is No Future for Online Safety Without Privacy and Security
itsfoss.com
Things I learnt about passkeys when building passkeybot
enzom.dev
Hybrid Aerial Underwater Drone – Bachelor Project [video]
youtube.com
Uplane (YC F25) Is Hiring Founding Engineers (Full-Stack and AI)
useparallel.com
Obviously cache snooping is going to be device-dependent, but the way this shows things happening is that if you write an address that exists in another CPU's cache, that the invalidation doesn't happen until that CPU later tries to read the address. That requires two snoop cycles be dedicated, which is a waste. The secondary CPU is already listening to the snoop bus, it can/should (and in practice does on basically all devices, AFAIK) be able to recognize that write fly by and free up its cache line preemptively.