diff options
| author | Paul Buetow <paul@buetow.org> | 2026-05-23 20:04:06 +0300 |
|---|---|---|
| committer | Paul Buetow <paul@buetow.org> | 2026-05-23 20:04:06 +0300 |
| commit | 4541ee21203f7c494530f142a6387244ac3a6c62 (patch) | |
| tree | 3279dcafffcbcb22662f01fea1f9afb0e4039663 /AGENTS.md | |
| parent | b4a172404be8b52ed62171dcbbd3e9f46e36ac67 (diff) | |
0c promote aggregate-only sampling defaults in raw output modes
Default aggregate-only sampling (rate 0) for futex* and clock_gettime
causes BPF to suppress ring-buffer events. In -plain, -flamegraph, and
headless -parquet modes there is no aggregate sink, so these probes
would emit no rows even when explicitly selected. Promote those defaults
to rate 1 during flag resolution; user-explicit overrides are preserved.
Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>
Diffstat (limited to 'AGENTS.md')
| -rw-r--r-- | AGENTS.md | 1 |
1 files changed, 1 insertions, 0 deletions
@@ -81,6 +81,7 @@ Generator source code: - **Sampling / aggregate-only mode**: - `-syscall-sampling-families` and `-syscall-sampling-syscalls` control per-family/per-syscall sampling (`0` = aggregate-only, `1` = all events, `N` = 1-in-N). - Current defaults include aggregate-only (`0`) for `futex`, `futex_wait`, `futex_wake`, `futex_requeue`, `futex_waitv`, and `clock_gettime`. + - In raw output modes (`-plain`, `-flamegraph`, headless `-parquet`) the default aggregate-only rates are automatically promoted to `1` because these modes lack a TUI aggregate sink. User-explicit `-syscall-sampling-syscalls` overrides are preserved. - **Additional metric dimensions**: - Address-space extent accumulator: `TotalAddressSpaceBytes` and `AddressSpaceBytesPerSec` in `statsengine.Snapshot`. - Per-event stream/export field `requested_sleep_ns` (from sleep tracepoints). |