Messages in this thread | | | From | "Perez-Gonzalez, Inaky" <> | Subject | migration_thread()'s priority too low? | Date | Fri, 28 Mar 2003 20:50:32 -0800 |
| |
Hi all, Ingo
I am having some trouble in the priority-inheritance wakeup code when using FIFO tasks - I was wondering: migration_thread() has the equivalent of a FIFO priority 0; thus, it will be left out by any FIFO task and migration won't work - I don't think this is causing the problem to my test cases, but I was curious anyway.
Setting it in 2.5.66's sched.c like this
@@ -2436,7 +2435,7 @@ */ static int migration_thread(void * data) { - struct sched_param param = { .sched_priority = MAX_RT_PRIO-1 }; + struct sched_param param = { .sched_priority = 0 }; int cpu = (long) data; runqueue_t *rq; int ret;
gives it max priority; it'd be interesting though to have an extra level so have FIFO 99 be 1 in the index and 0 still be free for system stuff.
Of course I can be missing anything really clear. Is it intentionate that migration_thread() has FIFO 0?
Iñaky Pérez-González -- Not speaking for Intel -- all opinions are my own (and my fault) - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/
| |