On Sat, Mar 21, 2026 at 05:10:02PM -0700, Andrew Morton wrote: > On Wed, 18 Mar 2026 18:45:17 +0000 Dmitry Ilvokhin <[email protected]> wrote: > > > The existing contention_begin/contention_end tracepoints fire on the > > waiter side. The lock holder's identity and stack can be captured at > > contention_begin time (e.g. perf lock contention --lock-owner), but > > this reflects the holder's state when a waiter arrives, not when the > > lock is actually released. > > > > This series adds a contended_release tracepoint that fires on the > > holder side when a lock with waiters is released. This provides: > > > > - Hold time estimation: when the holder's own acquisition was > > contended, its contention_end (acquisition) and contended_release > > can be correlated to measure how long the lock was held under > > contention. > > > > - The holder's stack at release time, which may differ from what perf lock > > contention --lock-owner captures if the holder does significant work > > between > > the waiter's arrival and the unlock. > > > > The series is structured as follows: > > > > 1. Remove unnecessary linux/sched.h include from trace/events/lock.h. > > 2. Extract __percpu_up_read() out of the inline percpu_up_read() to > > avoid binary size increase from adding a tracepoint. > > 3. Add contended_release tracepoint and instrument sleepable locks: > > mutex, rtmutex, semaphore, rwsem, percpu-rwsem, and rwbase_rt. > > AI review: > https://sashiko.dev/#/patchset/[email protected]
Thanks, Andrew, appreciate you sharing the link. The AI review looks reasonable. I'll go through it and address the feedback in the next revision. The kernel test robot is also reporting failures on some configs, which seem related to the Sashiko comments.
