[0/2] Refactor Logging Infrastructure

[yaakov-stein]

Stack:

• -> [0/2] Refactor Logging Infrastructure #485
• [1/2] Add flavor-specific log codegen callback #486
• [2/2] Add CGROUP_SOCK_ADDR Logging Infrastructure #491

Summary

This sets the stage for adding logging support for CGROUP_SOCK_ADDR. The first step is renaming the logging infrastructure in a way that isn't packet-specific (or making it clear that certain parts are packet-specific and naming accordingly), and adding a union that allows us to add the necessary fields to bf_log to support this change.

No functional change (besides the bf_log_type enum helper methods).

[github-actions]

Claude review of PR #485 (bbb165c)

Suggestions

• Add static_assert for struct bf_log size — src/libbpfilter/include/bpfilter/runtime.h:195 — struct grew from 112 to 120 bytes due to union layout; a static_assert would guard the ABI boundary and catch accidental regressions
• Zero ring buffer memory after reserve — src/libbpfilter/bpf/pkt_log.bpf.c:36 — bpf_ringbuf_reserve returns uninitialized memory; padding gaps and partially-filled header slices could leak kernel data to userspace (pre-existing, but natural to fix here)

Nits

• BF_LOG_TYPE_PACKET = 0 ambiguous with uninitialized data — src/libbpfilter/include/bpfilter/runtime.h:100 — starting the enum at 1 would distinguish valid entries from uninitialized ring buffer memory
• Group common-header writes separately from pkt fields — src/libbpfilter/bpf/pkt_log.bpf.c:33 — a blank line between common-header assignments and pkt-specific writes would mirror the struct layout separation

Workflow run