Stories by matt_d

Tessera: Unlocking Heterogeneous GPUs Through Kernel-Granularity Disaggregation

From SIMT to Systolic: A Foundation for GPU and TPU Architecture

Packrat Parsing at the Speed of Wasm [video]

Sparser, Faster, Lighter Transformer Language Models

When GPUs Fail Quietly: Observability-Aware Early Warning Beyond Telemetry

Stupid RCU Tricks: Corner-Case RCU Implementations

How Many Compilers Is Too Many? V8's History, Tradeoffs, and Architecture [video]

Fully-Automatic Type Inference for Borrows with Lifetimes

The GNU libc atanh is correctly rounded

MMU Handbook: Memory Management Units and TLBs

PEP 831 – Frame Pointers Everywhere: Enabling System-Level Observability

UpDown: Efficient Manycore based on Many Threading & Scalable Memory Parallelism

Reflections on 30 years of HPC programming

Recent lld/ELF performance improvements

Circuit Transformations, Loop Fusion, and Inductive Proof

AI for Systems: Using LLMs to Optimize Database Query Execution

Optimization of 32-bit Unsigned Division by Constants on 64-bit Targets

GCC Translation Validation Part 6: Uninitialized Memory

Agentic Code Optimization via Compiler-LLM Cooperation

Bespoke OLAP: Using AI to Synthesize Workload-Specific DBMS Engines from Scratch

9 10 11 12 13 14 15 16 17 18