beyond-pg-init: report PSI memory pressure to the host over vsock

[jaredLunde]

Why

The host memory controller (instd) right-sizes each VM's balloon/hotplug from a guest distress signal. Its sharpest signal is Linux PSI (/proc/pressure/memory), which measures memory-stall time directly — so it catches the buffered read() cache misses a database generates, which the previous major_faults signal misses. That's exactly the thrash a Postgres VM hits when its page cache is squeezed (a real incident: a Postgres VM pinned both vCPUs at 99% for 12h).

But this primitive never reported any guest resource stats over vsock, so the controller was flying blind on the workload that needs it most.

What

• Add a periodic (30s) GuestResourceStats (0xA2) report carrying PSI some.avg10 / full.avg10, multiplexed onto the existing substrate connection alongside heartbeats and log relay.
• PSI-only (omits disk_total_bytes) so the host skips disk billing for this report — Postgres disk usage isn't tracked here.
• Byte-compatible with instd's vsock_protocol::GuestResourceStatsPayload decoder. resource_stats_frame_is_stable pins the wire format the same way ready_frame_is_stable does, so it can't silently drift.
• A pure parse_memory_pressure() with a unit test; read_memory_pressure() returns None when PSI is unavailable.

Companion change

The host side (wire field, shared collector for the in-repo primitives, instd ingestion, and the controller treating PSI as the primary distress signal) lands in the beyond repo. instd sets psi=1 on the guest kernel cmdline (the kernel already ships CONFIG_PSI=y), so no kernel rebuild is needed. If PSI is unavailable the reporter sends nothing and the controller falls back to its balloon-stat signals.

Test

cargo test -p beyond-pg-init substrate — ready_frame_is_stable, resource_stats_frame_is_stable, parses_psi_memory all pass.

🤖 Generated with Claude Code