Messages in this thread | | | Date | Tue, 19 May 2015 15:57:47 +0900 | From | Namhyung Kim <> | Subject | Re: [PATCH 14/40] perf tools: Introduce machine__find*_thread_by_time() |
| |
On Mon, May 18, 2015 at 09:50:06PM +0200, Jiri Olsa wrote: > On Mon, May 18, 2015 at 09:30:29AM +0900, Namhyung Kim wrote: > > SNIP > > > +static struct thread * > > +__machine__findnew_thread_by_time(struct machine *machine, pid_t pid, pid_t tid, > > + u64 timestamp, bool create) > > +{ > > + struct thread *curr, *pos, *new; > > + struct thread *th = NULL; > > + struct rb_node **p; > > + struct rb_node *parent = NULL; > > + > > + if (!perf_has_index) > > + return ____machine__findnew_thread(machine, pid, tid, create); > > + > > + /* lookup current thread first */ > > + curr = ____machine__findnew_thread(machine, pid, tid, false); > > + if (curr && timestamp >= curr->start_time) > > + return curr; > > + > > + /* and then check dead threads tree & list */ > > + p = &machine->dead_threads.rb_node; > > + while (*p != NULL) { > > + parent = *p; > > + th = rb_entry(parent, struct thread, rb_node); > > + > > + if (th->tid == tid) { > > + list_for_each_entry(pos, &th->tid_node, tid_node) { > > + if (timestamp >= pos->start_time && > > + pos->start_time > th->start_time) { > > + th = pos; > > + break; > > + } > > hum, how do we know, there's not another thread on the list > fitting the timestamp >= pos->start_time condition as well?
Because the list is sorted on time - it'll see thread that has a larger start_time first so it can ensure that a thread which satisfies the expression first is the current thread at that time. Later threads in the list will satisfy the expression also but they're already dead.
Thanks, Namhyung
| |