Although rcutorture will occasionally do a 50-millisecond grace-period delay, these delays are quite rare. And rightly so, because otherwise the read rate would be quite low. Thie means that it can be important to identify whether or not a given run contained a long-delay read. This commit therefore inserts a trace_rcu_torture_read() event to flag runs containing long delays.
Showing the changes to include/trace (1 of 2 files). Show all
Unified · Split
| @@ -698,7 +698,10 @@ TRACE_EVENT(rcu_batch_end, · 697 unmodified lines | |||
| 698 | /* | 698 | /* |
| 699 | * Tracepoint for rcutorture readers. The first argument is the name | 699 | * Tracepoint for rcutorture readers. The first argument is the name |
| 700 | * of the RCU flavor from rcutorture's viewpoint and the second argument | 700 | * of the RCU flavor from rcutorture's viewpoint and the second argument |
| 701 | * is the callback address. | 701 | * is the callback address. The third argument is the start time in |
| 702 | * seconds, and the last two arguments are the grace period numbers | ||
| 703 | * at the beginning and end of the read, respectively. Note that the | ||
| 704 | * callback address can be NULL. | ||
| 702 | */ | 705 | */ |
| 703 | TRACE_EVENT(rcu_torture_read, | 706 | TRACE_EVENT(rcu_torture_read, |
| 704 | 707 | ||