[question] What would it take for off-cpu samples on Linux?

[lukehsiao]

samply is a sampling profiler and collects stack traces, per thread, at some sampling interval (the default 1000Hz, i.e. 1ms). On macOS, both on- and off-cpu samples are collected (so you can see under which stack you were blocking on a lock, for example). On Linux, only on-cpu samples are collected at the moment.

Huge fan of samply so far, but really would love off-cpu samples on Linux. Curious if there is a doc or discussion about what that would take to implement.