facebookincubator / dynolog

Dynolog is a telemetry daemon for performance monitoring and tracing. It exports metrics from different components in the system like the linux kernel, CPU, disks, Intel PT, GPUs etc. Dynolog also integrates with pytorch and can trigger traces for distributed training applications.
MIT License
188 stars 34 forks source link

fix empty cgroup path and cgroup output map cleanup issues #174

Closed Alston-Tang closed 10 months ago

Alston-Tang commented 10 months ago

Summary: 1) ignore tw tasks without a cgroup v2 path to avoid TwTaskMonitor opening bperf counters for a invalid cgroup

2) call disable() when BperfCountReader is destroyed so it can remove its inode key from the shared cgroup output map.

Differential Revision: D48657930

facebook-github-bot commented 10 months ago

This pull request was exported from Phabricator. Differential Revision: D48657930

facebook-github-bot commented 10 months ago

This pull request has been merged in facebookincubator/dynolog@68f5c4a93fa312232a0254ff4cd1d76c51494882.