From: Ingo Molnar Introduces two new /sys/block values: /sys/block/*/queue/max_hw_sectors_kb /sys/block/*/queue/max_sectors_kb max_hw_sectors_kb is the maximum that the driver can handle and is readonly. max_sectors_kb is the current max_sectors value and can be tuned by root. PAGE_SIZE granularity is enforced. It's all locking-safe and all affected layered drivers have been updated as well. The patch has been in testing for a couple of weeks already as part of the voluntary-preempt patches and it works just fine - people use it to reduce IDE IRQ handling latencies. Signed-off-by: Ingo Molnar Signed-off-by: Andrew Morton --- 25-akpm/drivers/block/ll_rw_blk.c | 70 +++++++++++++++++++++++++++++++++++--- 25-akpm/drivers/md/dm-table.c | 2 - 25-akpm/drivers/md/linear.c | 2 - 25-akpm/drivers/md/multipath.c | 4 +- 25-akpm/drivers/md/raid0.c | 2 - 25-akpm/drivers/md/raid1.c | 4 +- 25-akpm/include/linux/blkdev.h | 1 7 files changed, 74 insertions(+), 11 deletions(-) diff -puN drivers/block/ll_rw_blk.c~blk-max_sectors-tunables drivers/block/ll_rw_blk.c --- 25/drivers/block/ll_rw_blk.c~blk-max_sectors-tunables Wed Sep 8 14:23:46 2004 +++ 25-akpm/drivers/block/ll_rw_blk.c Wed Sep 8 14:23:46 2004 @@ -353,7 +353,7 @@ void blk_queue_max_sectors(request_queue printk("%s: set to minimum %d\n", __FUNCTION__, max_sectors); } - q->max_sectors = max_sectors; + q->max_sectors = q->max_hw_sectors = max_sectors; } EXPORT_SYMBOL(blk_queue_max_sectors); @@ -455,7 +455,8 @@ EXPORT_SYMBOL(blk_queue_hardsect_size); void blk_queue_stack_limits(request_queue_t *t, request_queue_t *b) { /* zero is "infinity" */ - t->max_sectors = min_not_zero(t->max_sectors,b->max_sectors); + t->max_sectors = t->max_hw_sectors = + min_not_zero(t->max_sectors,b->max_sectors); t->max_phys_segments = min(t->max_phys_segments,b->max_phys_segments); t->max_hw_segments = min(t->max_hw_segments,b->max_hw_segments); @@ -2600,11 +2601,11 @@ end_io: break; } - if (unlikely(bio_sectors(bio) > q->max_sectors)) { + if (unlikely(bio_sectors(bio) > q->max_hw_sectors)) { printk("bio too big device %s (%u > %u)\n", bdevname(bio->bi_bdev, b), bio_sectors(bio), - q->max_sectors); + q->max_hw_sectors); goto end_io; } @@ -3246,13 +3247,61 @@ queue_ra_store(struct request_queue *q, unsigned long ra_kb; ssize_t ret = queue_var_store(&ra_kb, page, count); + spin_lock_irq(q->queue_lock); if (ra_kb > (q->max_sectors >> 1)) ra_kb = (q->max_sectors >> 1); q->backing_dev_info.ra_pages = ra_kb >> (PAGE_CACHE_SHIFT - 10); + spin_unlock_irq(q->queue_lock); + + return ret; +} + +static ssize_t queue_max_sectors_show(struct request_queue *q, char *page) +{ + int max_sectors_kb = q->max_sectors >> 1; + + return queue_var_show(max_sectors_kb, (page)); +} + +static ssize_t +queue_max_sectors_store(struct request_queue *q, const char *page, size_t count) +{ + unsigned long max_sectors_kb, + max_hw_sectors_kb = q->max_hw_sectors >> 1, + page_kb = 1 << (PAGE_CACHE_SHIFT - 10); + ssize_t ret = queue_var_store(&max_sectors_kb, page, count); + int ra_kb; + + if (max_sectors_kb > max_hw_sectors_kb || max_sectors_kb < page_kb) + return -EINVAL; + /* + * Take the queue lock to update the readahead and max_sectors + * values synchronously: + */ + spin_lock_irq(q->queue_lock); + /* + * Trim readahead window as well, if necessary: + */ + ra_kb = q->backing_dev_info.ra_pages << (PAGE_CACHE_SHIFT - 10); + if (ra_kb > max_sectors_kb) + q->backing_dev_info.ra_pages = + max_sectors_kb >> (PAGE_CACHE_SHIFT - 10); + + q->max_sectors = max_sectors_kb << 1; + spin_unlock_irq(q->queue_lock); + return ret; } +static ssize_t queue_max_hw_sectors_show(struct request_queue *q, char *page) +{ + int max_hw_sectors_kb = q->max_hw_sectors >> 1; + + return queue_var_show(max_hw_sectors_kb, (page)); +} + + static struct queue_sysfs_entry queue_requests_entry = { .attr = {.name = "nr_requests", .mode = S_IRUGO | S_IWUSR }, .show = queue_requests_show, @@ -3265,9 +3314,22 @@ static struct queue_sysfs_entry queue_ra .store = queue_ra_store, }; +static struct queue_sysfs_entry queue_max_sectors_entry = { + .attr = {.name = "max_sectors_kb", .mode = S_IRUGO | S_IWUSR }, + .show = queue_max_sectors_show, + .store = queue_max_sectors_store, +}; + +static struct queue_sysfs_entry queue_max_hw_sectors_entry = { + .attr = {.name = "max_hw_sectors_kb", .mode = S_IRUGO }, + .show = queue_max_hw_sectors_show, +}; + static struct attribute *default_attrs[] = { &queue_requests_entry.attr, &queue_ra_entry.attr, + &queue_max_hw_sectors_entry.attr, + &queue_max_sectors_entry.attr, NULL, }; diff -puN drivers/md/dm-table.c~blk-max_sectors-tunables drivers/md/dm-table.c --- 25/drivers/md/dm-table.c~blk-max_sectors-tunables Wed Sep 8 14:23:46 2004 +++ 25-akpm/drivers/md/dm-table.c Wed Sep 8 14:23:46 2004 @@ -825,7 +825,7 @@ void dm_table_set_restrictions(struct dm * Make sure we obey the optimistic sub devices * restrictions. */ - q->max_sectors = t->limits.max_sectors; + blk_queue_max_sectors(q, t->limits.max_sectors); q->max_phys_segments = t->limits.max_phys_segments; q->max_hw_segments = t->limits.max_hw_segments; q->hardsect_size = t->limits.hardsect_size; diff -puN drivers/md/linear.c~blk-max_sectors-tunables drivers/md/linear.c --- 25/drivers/md/linear.c~blk-max_sectors-tunables Wed Sep 8 14:23:46 2004 +++ 25-akpm/drivers/md/linear.c Wed Sep 8 14:23:46 2004 @@ -154,7 +154,7 @@ static int linear_run (mddev_t *mddev) */ if (rdev->bdev->bd_disk->queue->merge_bvec_fn && mddev->queue->max_sectors > (PAGE_SIZE>>9)) - mddev->queue->max_sectors = (PAGE_SIZE>>9); + blk_queue_max_sectors(mddev->queue, PAGE_SIZE>>9); disk->size = rdev->size; mddev->array_size += rdev->size; diff -puN drivers/md/multipath.c~blk-max_sectors-tunables drivers/md/multipath.c --- 25/drivers/md/multipath.c~blk-max_sectors-tunables Wed Sep 8 14:23:46 2004 +++ 25-akpm/drivers/md/multipath.c Wed Sep 8 14:23:46 2004 @@ -333,7 +333,7 @@ static int multipath_add_disk(mddev_t *m */ if (rdev->bdev->bd_disk->queue->merge_bvec_fn && mddev->queue->max_sectors > (PAGE_SIZE>>9)) - mddev->queue->max_sectors = (PAGE_SIZE>>9); + blk_queue_max_sectors(mddev->queue, PAGE_SIZE>>9); conf->working_disks++; rdev->raid_disk = path; @@ -498,7 +498,7 @@ static int multipath_run (mddev_t *mddev * a merge_bvec_fn to be involved in multipath */ if (rdev->bdev->bd_disk->queue->merge_bvec_fn && mddev->queue->max_sectors > (PAGE_SIZE>>9)) - mddev->queue->max_sectors = (PAGE_SIZE>>9); + blk_queue_max_sectors(mddev->queue, PAGE_SIZE>>9); if (!rdev->faulty) conf->working_disks++; diff -puN drivers/md/raid0.c~blk-max_sectors-tunables drivers/md/raid0.c --- 25/drivers/md/raid0.c~blk-max_sectors-tunables Wed Sep 8 14:23:46 2004 +++ 25-akpm/drivers/md/raid0.c Wed Sep 8 14:23:46 2004 @@ -158,7 +158,7 @@ static int create_strip_zones (mddev_t * if (rdev1->bdev->bd_disk->queue->merge_bvec_fn && mddev->queue->max_sectors > (PAGE_SIZE>>9)) - mddev->queue->max_sectors = (PAGE_SIZE>>9); + blk_queue_max_sectors(mddev->queue, PAGE_SIZE>>9); if (!smallest || (rdev1->size size)) smallest = rdev1; diff -puN drivers/md/raid1.c~blk-max_sectors-tunables drivers/md/raid1.c --- 25/drivers/md/raid1.c~blk-max_sectors-tunables Wed Sep 8 14:23:46 2004 +++ 25-akpm/drivers/md/raid1.c Wed Sep 8 14:23:46 2004 @@ -752,7 +752,7 @@ static int raid1_add_disk(mddev_t *mddev */ if (rdev->bdev->bd_disk->queue->merge_bvec_fn && mddev->queue->max_sectors > (PAGE_SIZE>>9)) - mddev->queue->max_sectors = (PAGE_SIZE>>9); + blk_queue_max_sectors(mddev->queue, PAGE_SIZE>>9); p->head_position = 0; rdev->raid_disk = mirror; @@ -1198,7 +1198,7 @@ static int run(mddev_t *mddev) */ if (rdev->bdev->bd_disk->queue->merge_bvec_fn && mddev->queue->max_sectors > (PAGE_SIZE>>9)) - mddev->queue->max_sectors = (PAGE_SIZE>>9); + blk_queue_max_sectors(mddev->queue, PAGE_SIZE>>9); disk->head_position = 0; if (!rdev->faulty && rdev->in_sync) diff -puN include/linux/blkdev.h~blk-max_sectors-tunables include/linux/blkdev.h --- 25/include/linux/blkdev.h~blk-max_sectors-tunables Wed Sep 8 14:23:46 2004 +++ 25-akpm/include/linux/blkdev.h Wed Sep 8 14:23:46 2004 @@ -362,6 +362,7 @@ struct request_queue unsigned int nr_batching; unsigned short max_sectors; + unsigned short max_hw_sectors; unsigned short max_phys_segments; unsigned short max_hw_segments; unsigned short hardsect_size; _