Jeff Moyer (1): block: trace event block fix unassigned field Jens Axboe (2): block: add internal hd part table references Merge branch 'for-2.6.38/core' into for-next Sedat Dilek (1): Merge branch 'for-next' of git://git.kernel.org/.../axboe/linux-2.6-block Shaohua Li (2): block cfq: don't use atomic_t for cfq_queue block cfq: don't use atomic_t for cfq_group block/blk-core.c | 6 ++-- block/blk-merge.c | 2 +- block/cfq-iosched.c | 50 +++++++++++++++++++++++------------------ block/genhd.c | 3 +- drivers/md/dm.c | 2 +- fs/partitions/check.c | 8 ++---- include/linux/genhd.h | 27 +++++++++++++++++++++- include/linux/kref.h | 1 - include/trace/events/block.h | 6 +++- lib/kref.c | 12 ---------- 10 files changed, 67 insertions(+), 50 deletions(-) diff --git a/block/blk-core.c b/block/blk-core.c index 500c080..2f4002f 100644 --- a/block/blk-core.c +++ b/block/blk-core.c @@ -70,7 +70,7 @@ static void drive_stat_acct(struct request *rq, int new_io) part_stat_inc(cpu, part, merges[rw]); } else { part = disk_map_sector_rcu(rq->rq_disk, blk_rq_pos(rq)); - if (!kref_test_and_get(&part->ref)) { + if (!hd_struct_try_get(part)) { /* * The partition is already being removed, * the request will be accounted on the disk only @@ -80,7 +80,7 @@ static void drive_stat_acct(struct request *rq, int new_io) * it as any other partition. */ part = &rq->rq_disk->part0; - kref_get(&part->ref); + hd_struct_get(part); } part_round_stats(cpu, part); part_inc_in_flight(part, rw); @@ -1818,7 +1818,7 @@ static void blk_account_io_done(struct request *req) part_round_stats(cpu, part); part_dec_in_flight(part, rw); - kref_put(&part->ref, __delete_partition); + hd_struct_put(part); part_stat_unlock(); } } diff --git a/block/blk-merge.c b/block/blk-merge.c index 23ea74b..ea85e20 100644 --- a/block/blk-merge.c +++ b/block/blk-merge.c @@ -356,7 +356,7 @@ static void blk_account_io_merge(struct request *req) part_round_stats(cpu, part); part_dec_in_flight(part, rq_data_dir(req)); - kref_put(&part->ref, __delete_partition); + hd_struct_put(part); part_stat_unlock(); } } diff --git a/block/cfq-iosched.c b/block/cfq-iosched.c index 1ac20f5..8427697 100644 --- a/block/cfq-iosched.c +++ b/block/cfq-iosched.c @@ -96,7 +96,7 @@ struct cfq_rb_root { */ struct cfq_queue { /* reference count */ - atomic_t ref; + int ref; /* various state flags, see below */ unsigned int flags; /* parent cfq_data */ @@ -207,7 +207,7 @@ struct cfq_group { struct blkio_group blkg; #ifdef CONFIG_CFQ_GROUP_IOSCHED struct hlist_node cfqd_node; - atomic_t ref; + int ref; #endif /* number of requests that are on the dispatch list or inside driver */ int dispatched; @@ -1014,7 +1014,7 @@ cfq_find_alloc_cfqg(struct cfq_data *cfqd, struct cgroup *cgroup, int create) * elevator which will be dropped by either elevator exit * or cgroup deletion path depending on who is exiting first. */ - atomic_set(&cfqg->ref, 1); + cfqg->ref = 1; /* * Add group onto cgroup list. It might happen that bdi->dev is @@ -1059,7 +1059,7 @@ static struct cfq_group *cfq_get_cfqg(struct cfq_data *cfqd, int create) static inline struct cfq_group *cfq_ref_get_cfqg(struct cfq_group *cfqg) { - atomic_inc(&cfqg->ref); + cfqg->ref++; return cfqg; } @@ -1071,7 +1071,7 @@ static void cfq_link_cfqq_cfqg(struct cfq_queue *cfqq, struct cfq_group *cfqg) cfqq->cfqg = cfqg; /* cfqq reference on cfqg */ - atomic_inc(&cfqq->cfqg->ref); + cfqq->cfqg->ref++; } static void cfq_put_cfqg(struct cfq_group *cfqg) @@ -1079,8 +1079,9 @@ static void cfq_put_cfqg(struct cfq_group *cfqg) struct cfq_rb_root *st; int i, j; - BUG_ON(atomic_read(&cfqg->ref) <= 0); - if (!atomic_dec_and_test(&cfqg->ref)) + BUG_ON(cfqg->ref <= 0); + cfqg->ref--; + if (cfqg->ref) return; for_each_cfqg_st(cfqg, i, j, st) BUG_ON(!RB_EMPTY_ROOT(&st->rb)); @@ -1188,7 +1189,7 @@ static void cfq_service_tree_add(struct cfq_data *cfqd, struct cfq_queue *cfqq, cfq_group_service_tree_del(cfqd, cfqq->cfqg); cfqq->orig_cfqg = cfqq->cfqg; cfqq->cfqg = &cfqd->root_group; - atomic_inc(&cfqd->root_group.ref); + cfqd->root_group.ref++; group_changed = 1; } else if (!cfqd->cfq_group_isolation && cfqq_type(cfqq) == SYNC_WORKLOAD && cfqq->orig_cfqg) { @@ -2025,7 +2026,7 @@ static int cfqq_process_refs(struct cfq_queue *cfqq) int process_refs, io_refs; io_refs = cfqq->allocated[READ] + cfqq->allocated[WRITE]; - process_refs = atomic_read(&cfqq->ref) - io_refs; + process_refs = cfqq->ref - io_refs; BUG_ON(process_refs < 0); return process_refs; } @@ -2065,10 +2066,10 @@ static void cfq_setup_merge(struct cfq_queue *cfqq, struct cfq_queue *new_cfqq) */ if (new_process_refs >= process_refs) { cfqq->new_cfqq = new_cfqq; - atomic_add(process_refs, &new_cfqq->ref); + new_cfqq->ref += process_refs; } else { new_cfqq->new_cfqq = cfqq; - atomic_add(new_process_refs, &cfqq->ref); + cfqq->ref += new_process_refs; } } @@ -2532,9 +2533,10 @@ static void cfq_put_queue(struct cfq_queue *cfqq) struct cfq_data *cfqd = cfqq->cfqd; struct cfq_group *cfqg, *orig_cfqg; - BUG_ON(atomic_read(&cfqq->ref) <= 0); + BUG_ON(cfqq->ref <= 0); - if (!atomic_dec_and_test(&cfqq->ref)) + cfqq->ref--; + if (cfqq->ref) return; cfq_log_cfqq(cfqd, cfqq, "put_queue"); @@ -2837,7 +2839,7 @@ static void cfq_init_cfqq(struct cfq_data *cfqd, struct cfq_queue *cfqq, RB_CLEAR_NODE(&cfqq->p_node); INIT_LIST_HEAD(&cfqq->fifo); - atomic_set(&cfqq->ref, 0); + cfqq->ref = 0; cfqq->cfqd = cfqd; cfq_mark_cfqq_prio_changed(cfqq); @@ -2973,11 +2975,11 @@ cfq_get_queue(struct cfq_data *cfqd, bool is_sync, struct io_context *ioc, * pin the queue now that it's allocated, scheduler exit will prune it */ if (!is_sync && !(*async_cfqq)) { - atomic_inc(&cfqq->ref); + cfqq->ref++; *async_cfqq = cfqq; } - atomic_inc(&cfqq->ref); + cfqq->ref++; return cfqq; } @@ -3679,13 +3681,13 @@ new_queue: } cfqq->allocated[rw]++; - atomic_inc(&cfqq->ref); - - spin_unlock_irqrestore(q->queue_lock, flags); - + cfqq->ref++; rq->elevator_private = cic; rq->elevator_private2 = cfqq; rq->elevator_private3 = cfq_ref_get_cfqg(cfqq->cfqg); + + spin_unlock_irqrestore(q->queue_lock, flags); + return 0; queue_fail: @@ -3860,6 +3862,10 @@ static void *cfq_init_queue(struct request_queue *q) if (!cfqd) return NULL; + /* + * Don't need take queue_lock in the routine, since we are + * initializing the ioscheduler, and nobody is using cfqd + */ cfqd->cic_index = i; /* Init root service tree */ @@ -3879,7 +3885,7 @@ static void *cfq_init_queue(struct request_queue *q) * Take a reference to root group which we never drop. This is just * to make sure that cfq_put_cfqg() does not try to kfree root group */ - atomic_set(&cfqg->ref, 1); + cfqg->ref = 1; rcu_read_lock(); cfq_blkiocg_add_blkio_group(&blkio_root_cgroup, &cfqg->blkg, (void *)cfqd, 0); @@ -3899,7 +3905,7 @@ static void *cfq_init_queue(struct request_queue *q) * will not attempt to free it. */ cfq_init_cfqq(cfqd, &cfqd->oom_cfqq, 1, 0); - atomic_inc(&cfqd->oom_cfqq.ref); + cfqd->oom_cfqq.ref++; cfq_link_cfqq_cfqg(&cfqd->oom_cfqq, &cfqd->root_group); INIT_LIST_HEAD(&cfqd->cic_list); diff --git a/block/genhd.c b/block/genhd.c index 0093ddfa..6a5b772 100644 --- a/block/genhd.c +++ b/block/genhd.c @@ -1263,7 +1263,8 @@ struct gendisk *alloc_disk_node(int minors, int node_id) return NULL; } disk->part_tbl->part[0] = &disk->part0; - kref_init(&disk->part0.ref); + + hd_ref_init(&disk->part0); disk->minors = minors; rand_initialize_disk(disk); diff --git a/drivers/md/dm.c b/drivers/md/dm.c index 0a3ee08..0de6921 100644 --- a/drivers/md/dm.c +++ b/drivers/md/dm.c @@ -629,7 +629,7 @@ static void dec_pending(struct dm_io *io, int error) queue_io(md, bio); } else { /* done with normal IO or empty flush */ - trace_block_bio_complete(md->queue, bio); + trace_block_bio_complete(md->queue, bio, io_error); bio_endio(bio, io_error); } } diff --git a/fs/partitions/check.c b/fs/partitions/check.c index c00b12d..9c21119 100644 --- a/fs/partitions/check.c +++ b/fs/partitions/check.c @@ -381,10 +381,8 @@ static void delete_partition_rcu_cb(struct rcu_head *head) put_device(part_to_dev(part)); } -void __delete_partition(struct kref *ref) +void __delete_partition(struct hd_struct *part) { - struct hd_struct *part = container_of(ref, struct hd_struct, ref); - call_rcu(&part->rcu_head, delete_partition_rcu_cb); } @@ -406,7 +404,7 @@ void delete_partition(struct gendisk *disk, int partno) kobject_put(part->holder_dir); device_del(part_to_dev(part)); - kref_put(&part->ref, __delete_partition); + hd_struct_put(part); } static ssize_t whole_disk_show(struct device *dev, @@ -505,7 +503,7 @@ struct hd_struct *add_partition(struct gendisk *disk, int partno, if (!dev_get_uevent_suppress(ddev)) kobject_uevent(&pdev->kobj, KOBJ_ADD); - kref_init(&p->ref); + hd_ref_init(p); return p; out_free_info: diff --git a/include/linux/genhd.h b/include/linux/genhd.h index 84b49a2..c0d5f69 100644 --- a/include/linux/genhd.h +++ b/include/linux/genhd.h @@ -115,8 +115,8 @@ struct hd_struct { #else struct disk_stats dkstats; #endif + atomic_t ref; struct rcu_head rcu_head; - struct kref ref; }; #define GENHD_FL_REMOVABLE 1 @@ -598,7 +598,7 @@ extern struct hd_struct * __must_check add_partition(struct gendisk *disk, sector_t len, int flags, struct partition_meta_info *info); -extern void __delete_partition(struct kref *ref); +extern void __delete_partition(struct hd_struct *); extern void delete_partition(struct gendisk *, int); extern void printk_all_partitions(void); @@ -627,6 +627,29 @@ extern ssize_t part_fail_store(struct device *dev, const char *buf, size_t count); #endif /* CONFIG_FAIL_MAKE_REQUEST */ +static inline void hd_ref_init(struct hd_struct *part) +{ + atomic_set(&part->ref, 1); + smp_mb(); +} + +static inline void hd_struct_get(struct hd_struct *part) +{ + atomic_inc(&part->ref); + smp_mb__after_atomic_inc(); +} + +static inline int hd_struct_try_get(struct hd_struct *part) +{ + return atomic_inc_not_zero(&part->ref); +} + +static inline void hd_struct_put(struct hd_struct *part) +{ + if (atomic_dec_and_test(&part->ref)) + __delete_partition(part); +} + #else /* CONFIG_BLOCK */ static inline void printk_all_partitions(void) { } diff --git a/include/linux/kref.h b/include/linux/kref.h index 04f194a..d4a62ab 100644 --- a/include/linux/kref.h +++ b/include/linux/kref.h @@ -23,7 +23,6 @@ struct kref { void kref_init(struct kref *kref); void kref_get(struct kref *kref); -int kref_test_and_get(struct kref *kref); int kref_put(struct kref *kref, void (*release) (struct kref *kref)); int kref_sub(struct kref *kref, unsigned int count, void (*release) (struct kref *kref)); diff --git a/include/trace/events/block.h b/include/trace/events/block.h index b56c65d..aba421d 100644 --- a/include/trace/events/block.h +++ b/include/trace/events/block.h @@ -206,15 +206,16 @@ TRACE_EVENT(block_bio_bounce, * block_bio_complete - completed all work on the block operation * @q: queue holding the block operation * @bio: block operation completed + * @error: io error value * * This tracepoint indicates there is no further work to do on this * block IO operation @bio. */ TRACE_EVENT(block_bio_complete, - TP_PROTO(struct request_queue *q, struct bio *bio), + TP_PROTO(struct request_queue *q, struct bio *bio, int error), - TP_ARGS(q, bio), + TP_ARGS(q, bio, error), TP_STRUCT__entry( __field( dev_t, dev ) @@ -228,6 +229,7 @@ TRACE_EVENT(block_bio_complete, __entry->dev = bio->bi_bdev->bd_dev; __entry->sector = bio->bi_sector; __entry->nr_sector = bio->bi_size >> 9; + __entry->error = error; blk_fill_rwbs(__entry->rwbs, bio->bi_rw, bio->bi_size); ), diff --git a/lib/kref.c b/lib/kref.c index 52d0831..3efb882 100644 --- a/lib/kref.c +++ b/lib/kref.c @@ -37,18 +37,6 @@ void kref_get(struct kref *kref) } /** - * kref_test_and_get - increment refcount for object only if refcount is not - * zero. - * @kref: object. - * - * Return non-zero if the refcount was incremented, 0 otherwise - */ -int kref_test_and_get(struct kref *kref) -{ - return atomic_inc_not_zero(&kref->refcount); -} - -/** * kref_put - decrement refcount for object. * @kref: object. * @release: pointer to the function that will clean up the object when the