Messages in this thread | | | Date | Wed, 16 Dec 1998 21:29:51 +0100 (CET) | From | Andrea Arcangeli <> | Subject | Re: idle task was __cliing the local CPU even if not needed |
| |
On Tue, 15 Dec 1998, Andrea Arcangeli wrote:
>With this patch applyed the idle task run schedule() only if need_resched
I had a bit more of time to test the thing and seems that the performance are a bit worse?!?! I don' t understand why...
I was also thinking to reduce the idle task to a for(;;) asm("hlt"); and learn to reschedule_idle how do the task switch (and set the future next idle task eip properly) if the current process is the idle task. This should improve performance no? But according to me also the last patch should have improved and instead it made things a bit worse... Probably I am missing something of important...
- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.rutgers.edu Please read the FAQ at http://www.tux.org/lkml/
| |