BFS 420: cleanup in tick handling

The cpu on stack is not needed, so remove it.

kernel/sched/bfs.c +1 -4

@@ -2822,8 +2822,7 @@ void wake_up_idle_cpu(int cpu);
   */
  void scheduler_tick(void)
  {
-	int cpu __maybe_unused = smp_processor_id();
-	struct rq *rq = cpu_rq(cpu);
+	struct rq *rq = this_rq();
 
  	sched_clock_tick();
  	/* grq lock not grabbed, so only update rq clock */
--
--

To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to [email protected]
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/

Hillf, I noticed you posting a few patches to BFS recently. Just wanted to make you aware that there is a certain erm communication issue between the

kernel maintainers and the author of BFS. Since he is busy in real life you should contact him directly or cc him. See tail -1

Documentation/scheduler/sched-BFS.txt for email.

Actually the lowest deadline task selection algorithm now BFS used is O(n). I have already had an idea to enhance the lowest deadline task selection algorithm to O(1)

Quoted 8 lines · 在 2012-5-22 下午9:50，"Hillf Danton" <[email protected]>写道：

Quoted 4 lines · Chen

Actually the lowest deadline task selection algorithm now BFS used is O(n).
I have already had an idea to enhance the lowest deadline task selection algorithm to O(1)

Would you please announce it on LKML?

First of all let me talk how to do it .We can use 40 list_head instead of the single list for time sharing scheduling policy.
Then for task selection we will select the task from these 40 list_head(actually not 40, just decide by next_sched_bit), then make a comparsion of these tasks(actually in the worse case we just need to do 40 times of comparsion) and pick the lowest task. If a task wake, it would be the first node of (queue + p->prio). If a task run out of its timeslice, it will become the last node of(queue + p->prio). This is a concept of BFS deadline presort algorithm and the time complexity is O(1)

Quoted 28 lines · 在 2012-5-22 下午10:04，"Hillf Danton" <[email protected]>写道：

Quoted 25 lines · Chen

Quoted 9 lines · 在 2012-5-22 下午9:50，"Hillf Danton" <[email protected]>写道：

Quoted 5 lines · Chen

Actually the lowest deadline task selection algorithm now BFS used is O(n).
I have already had an idea to enhance the lowest deadline task selection
algorithm to O(1)

Would you please announce it on LKML?

First of all let me talk how to do it .We can use 40 list_head instead of
the single list for time sharing scheduling policy.
Then for task selection we will select the task from these 40 list_head(actually not 40, just decide by next_sched_bit), then make a comparsion of these tasks(actually in the worse case we just need to do 40
times of comparsion) and pick the lowest task. If a task wake, it would be
the first node of (queue + p->prio). If a task run out of its timeslice, it
will become the last node of(queue + p->prio). This is a concept of BFS deadline presort algorithm and the time complexity is O(1)

Quoted 1 line

Thank you for shedding light, I will study it soon.

If you are free could you also help me to maintain my cpu scheduler, thanks. . :-)

Quoted 7 lines · 在 2012-5-22 下午9:50，"Hillf Danton" <[email protected]>写道：

Quoted 4 lines · Chen

Actually the lowest deadline task selection algorithm now BFS used is O(n).
I have already had an idea to enhance the lowest deadline task selection algorithm to O(1)

Quoted 1 line

Would you please announce it on LKML?

First of all let me talk how to do it .We can use 40 list_head instead of the single list for time sharing scheduling policy.
Then for task selection we will select the task from these 40 list_head(actually not 40, just decide by next_sched_bit), then make a comparsion of these tasks(actually in the worse case we just need to do 40 times of comparsion) and pick the lowest task.

If a task wake, it would be
the first node of (queue + p->prio). If a task run out of its timeslice, it will become the last node of(queue + p->prio). This is a concept of BFS deadline presort algorithm and the time complexity is O(1)