Messages in this thread Patch in this message | | | From | Charles Baylis <> | Subject | [RFC] [PATCH] use nice values in deadline IO scheduler | Date | Wed, 11 Dec 2002 15:44:16 +0000 |
| |
This untested patch uses the nice value of the current task to scale the deadline for new read requests.
Does current contain a pointer to the task which caused the IO request at this point? Is there any other reason why this might be a daft thing to do?
--- drivers/block/deadline-iosched.c~std 2002-12-11 14:33:48.000000000 +0000 +++ drivers/block/deadline-iosched.c 2002-12-11 15:17:58.000000000 +0000 @@ -17,6 +17,7 @@ #include <linux/init.h> #include <linux/compiler.h> #include <linux/hash.h> +#include <linux/sched.h> /* * feel free to try other values :-). read_expire value is the timeout for @@ -81,6 +82,19 @@ static kmem_cache_t *drq_pool; #define RQ_DATA(rq) ((struct deadline_rq *) (rq)->elevator_private) /* + * scale_deadline + */ +static int scale_deadline(int default_deadline) +{ + int prio = current->static_prio - MAX_RT_PRIO; + /* make priorities higher than nice -10 equal to nice -10 */ + if (prio < 10) + prio = 10; + /* scale the deadline according to priority */ + return default_deadline * prio/20; +} + +/* * rq hash */ static inline void __deadline_del_rq_hash(struct deadline_rq *drq) @@ -440,7 +454,7 @@ deadline_add_request(request_queue_t *q, /* * set expire time and add to fifo list */ - drq->expires = jiffies + dd->read_expire; + drq->expires = jiffies + scale_deadline(dd->read_expire); list_add_tail(&drq->fifo, &dd->read_fifo); } } - 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/
| |