JR · @dniminenn
I build things that have to be measurably correct.
Notes on precision timing, audio, embedded systems, and debugging by measurement. Most of these started as an idle question and ran further than they had any right to.
Precision timing
GPS-disciplined clocks, NTP and PTP, nanosecond budgets, and proving a clock is honest.
Audio, end to end
Bit-perfect playback paths, DSP, LED visualizers, and chasing latency you can hear.
Embedded systems
ESP32, RP2040, and the firmware that has to keep running when the network does not.
Systems debugging
Reading the kernel, tracing the wire, measuring instead of guessing.
Selected writing
Putting back the bass that was never recorded
I prepped a music set for a 1500-watt DIY rig and synthesized a missing sub-bass octave into tracks rolled off below 40 Hz. Measure first, fix only what the numbers justify, don't touch anything else.
What is putting those bytes on the wire?
The storage interface on a production cluster was pulling 4 to 12x more than the public interface where client requests arrive. Getting to a defensible number meant distrusting everything, including my own profiler, which was off by 30x.
A Dovecot optimization the compiler had been deleting all along
An IMAP THREAD command was panicking inside Dovecot's array code. The crash was real, but the comment above it described an optimization that hadn't run in any -O2 build ever shipped. ATTR_PURE turned it into dead code, and the same annotation decided which line showed up in the backtrace.
A Ceph cluster I throw away on every reboot
I wanted a real Ceph endpoint (S3, CephFS, RBD) to test app code against, without a k8s cluster or a single byte hitting my SSD. One container, ~30 seconds, all in RAM. The version sweep is where it got interesting.