Messages in this thread | | | From | "Davide Libenzi" <> | Subject | Re: New schedule() and semaphore implementation ... | Date | Fri, 11 Jun 1999 18:43:42 +0200 |
| |
Hi andrea,
>Apologies, I overlooked your patch.
I'm happy !!
>I suggest also to search on the list a patch that avoids recalculating >counters on all tasks. It allows you to recalculate the counters only on >the runqueue. So you could run gds_remove_task();gds_add_task() >unconditionally in gds_switch().
I don't know if this is correct because of the new version of "up". With the actual code waiting processes counter were incremented taking a chance to exit from wait queue. Anyway I agree with the fact that it will be faster.
>I suggest you also to move the >slot-fields near the runqueue fields, so you may get them on the same >cacheline while inserting/removing the task from the runqueue.
100 % Agree.
Cheers, Davide.
-- "Debian, the Freedom in Freedom"
- 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/
| |