#MQTT

Selecting a firmware stack for a memory-constrained microcontroller is usually guided by aggregate binary sizes, which report how much SRAM a build consumes but not where it goes or why. We address that gap with a per-symbol attribution method that decomposes static RAM into architectural categories — scheduler heap, network buffers and pools, task stacks, and linker-reserved regions — by bounding sized ELF symbols to the .data, .bss and .stack section ranges. The method is validated by reconciling every category sum exactly against the size-based totals for all sixteen builds. We apply it to eight cumulative firmware milestones, from a bare task scaffold to a full MQTT-over-Ethernet client, implemented twice — in FreeRTOS-C and in Embassy-Rust — on the CH32V203RBT6, a RISC-V microcontroller with 64 KB SRAM and an on-chip Ethernet MAC and PHY, with functional parity verified on hardware at every milestone. The decomposition shows that both stacks spend a nearly identical share of their RAM on networking (63.8% versus 63.9%), yet the absolute cost differs by 4.4 times: 40,719 bytes for lwIP against 9,316 bytes for smoltcp. FreeRTOS task stacks grow linearly at roughly 1 KB per task while Embassy async state grows with state-machine complexity. We also report that lwIP could not be integrated under an all-static discipline on this platform, requiring a 20 KB arena that smoltcp does not. The measurement harness, the attribution tool, and the sixteen firmware implementations are released as artefacts.

Opening a broker to browsers and devices on the internet is less about encryption than identity. Each entrance now accepts one kind of credential, every token is bound to its own name and addressed to this broker, and TLS ends in nginx.

Each project began with its interface served from inside the borrowed Colab runtime, through Colab's own proxy — until Colab warned against it. Moving the interface out made the runtime a pure worker, and left two codebases held together by nothing but an agreement.

A laptop and a borrowed Colab GPU talked through a public MQTT broker that anyone could read. Giving that channel an identity took an afternoon on the broker and eight obstacles in the clients — mostly safeguards that were right for the old setup and wrong for the new one.

A Colab runtime has no address anyone can call, so a laptop and a borrowed GPU talk through a broker both of them dial out to. Sending commands was the easy half. Knowing — whether the worker is alive, whether a command arrived once — is where six silent failures hide.