diff options
author | Tejun Heo <tj@kernel.org> | 2011-12-14 00:33:42 +0100 |
---|---|---|
committer | Jens Axboe <axboe@kernel.dk> | 2011-12-14 00:33:42 +0100 |
commit | f1f8cc94651738b418ba54c039df536303b91704 (patch) | |
tree | eb8bc5a33dec104ab32a935a5bb1e1da2e7cdd34 /block/cfq-iosched.c | |
parent | block, cfq: restructure io_cq creation path for io_context interface cleanup (diff) | |
download | linux-f1f8cc94651738b418ba54c039df536303b91704.tar.xz linux-f1f8cc94651738b418ba54c039df536303b91704.zip |
block, cfq: move icq creation and rq->elv.icq association to block core
Now block layer knows everything necessary to create and associate
icq's with requests. Move ioc_create_icq() to blk-ioc.c and update
get_request() such that, if elevator_type->icq_size is set, requests
are automatically associated with their matching icq's before
elv_set_request(). io_context reference is also managed by block core
on request alloc/free.
* Only ioprio/cgroup changed handling remains from cfq_get_cic().
Collapsed into cfq_set_request().
* This removes queue kicking on icq allocation failure (for now). As
icq allocation failure is rare and the only effect of queue kicking
achieved was possibily accelerating queue processing, this change
shouldn't be noticeable.
There is a larger underlying problem. Unlike request allocation,
icq allocation is not guaranteed to succeed eventually after
retries. The number of icq is unbound and thus mempool can't be the
solution either. This effectively adds allocation dependency on
memory free path and thus possibility of deadlock.
This usually wouldn't happen because icq allocation is not a hot
path and, even when the condition triggers, it's highly unlikely
that none of the writeback workers already has icq.
However, this is still possible especially if elevator is being
switched under high memory pressure, so we better get it fixed.
Probably the only solution is just bypassing elevator and appending
to dispatch queue on any elevator allocation failure.
* Comment added to explain how icq's are managed and synchronized.
This completes cleanup of io_context interface.
Signed-off-by: Tejun Heo <tj@kernel.org>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Diffstat (limited to 'block/cfq-iosched.c')
-rw-r--r-- | block/cfq-iosched.c | 135 |
1 files changed, 11 insertions, 124 deletions
diff --git a/block/cfq-iosched.c b/block/cfq-iosched.c index 11f49d036845..f3b44c394e6d 100644 --- a/block/cfq-iosched.c +++ b/block/cfq-iosched.c @@ -2935,117 +2935,6 @@ cfq_get_queue(struct cfq_data *cfqd, bool is_sync, struct io_context *ioc, return cfqq; } -/** - * ioc_create_icq - create and link io_cq - * @q: request_queue of interest - * @gfp_mask: allocation mask - * - * Make sure io_cq linking %current->io_context and @q exists. If either - * io_context and/or icq don't exist, they will be created using @gfp_mask. - * - * The caller is responsible for ensuring @ioc won't go away and @q is - * alive and will stay alive until this function returns. - */ -static struct io_cq *ioc_create_icq(struct request_queue *q, gfp_t gfp_mask) -{ - struct elevator_type *et = q->elevator->type; - struct io_context *ioc; - struct io_cq *icq; - - /* allocate stuff */ - ioc = create_io_context(current, gfp_mask, q->node); - if (!ioc) - return NULL; - - icq = kmem_cache_alloc_node(et->icq_cache, gfp_mask | __GFP_ZERO, - q->node); - if (!icq) - return NULL; - - if (radix_tree_preload(gfp_mask) < 0) { - kmem_cache_free(et->icq_cache, icq); - return NULL; - } - - icq->ioc = ioc; - icq->q = q; - INIT_LIST_HEAD(&icq->q_node); - INIT_HLIST_NODE(&icq->ioc_node); - - /* lock both q and ioc and try to link @icq */ - spin_lock_irq(q->queue_lock); - spin_lock(&ioc->lock); - - if (likely(!radix_tree_insert(&ioc->icq_tree, q->id, icq))) { - hlist_add_head(&icq->ioc_node, &ioc->icq_list); - list_add(&icq->q_node, &q->icq_list); - if (et->ops.elevator_init_icq_fn) - et->ops.elevator_init_icq_fn(icq); - } else { - kmem_cache_free(et->icq_cache, icq); - icq = ioc_lookup_icq(ioc, q); - if (!icq) - printk(KERN_ERR "cfq: icq link failed!\n"); - } - - spin_unlock(&ioc->lock); - spin_unlock_irq(q->queue_lock); - radix_tree_preload_end(); - return icq; -} - -/** - * cfq_get_cic - acquire cfq_io_cq and bump refcnt on io_context - * @cfqd: cfqd to setup cic for - * @gfp_mask: allocation mask - * - * Return cfq_io_cq associating @cfqd and %current->io_context and - * bump refcnt on io_context. If ioc or cic doesn't exist, they're created - * using @gfp_mask. - * - * Must be called under queue_lock which may be released and re-acquired. - * This function also may sleep depending on @gfp_mask. - */ -static struct cfq_io_cq *cfq_get_cic(struct cfq_data *cfqd, gfp_t gfp_mask) -{ - struct request_queue *q = cfqd->queue; - struct cfq_io_cq *cic = NULL; - struct io_context *ioc; - - lockdep_assert_held(q->queue_lock); - - while (true) { - /* fast path */ - ioc = current->io_context; - if (likely(ioc)) { - cic = cfq_cic_lookup(cfqd, ioc); - if (likely(cic)) - break; - } - - /* slow path - unlock, create missing ones and retry */ - spin_unlock_irq(q->queue_lock); - cic = icq_to_cic(ioc_create_icq(q, gfp_mask)); - spin_lock_irq(q->queue_lock); - if (!cic) - return NULL; - } - - /* bump @ioc's refcnt and handle changed notifications */ - get_io_context(ioc); - - if (unlikely(cic->icq.changed)) { - if (test_and_clear_bit(ICQ_IOPRIO_CHANGED, &cic->icq.changed)) - changed_ioprio(cic); -#ifdef CONFIG_CFQ_GROUP_IOSCHED - if (test_and_clear_bit(ICQ_CGROUP_CHANGED, &cic->icq.changed)) - changed_cgroup(cic); -#endif - } - - return cic; -} - static void __cfq_update_io_thinktime(struct cfq_ttime *ttime, unsigned long slice_idle) { @@ -3524,8 +3413,6 @@ static void cfq_put_request(struct request *rq) BUG_ON(!cfqq->allocated[rw]); cfqq->allocated[rw]--; - put_io_context(RQ_CIC(rq)->icq.ioc, cfqq->cfqd->queue); - /* Put down rq reference on cfqg */ cfq_put_cfqg(RQ_CFQG(rq)); rq->elv.priv[0] = NULL; @@ -3574,7 +3461,7 @@ static int cfq_set_request(struct request_queue *q, struct request *rq, gfp_t gfp_mask) { struct cfq_data *cfqd = q->elevator->elevator_data; - struct cfq_io_cq *cic; + struct cfq_io_cq *cic = icq_to_cic(rq->elv.icq); const int rw = rq_data_dir(rq); const bool is_sync = rq_is_sync(rq); struct cfq_queue *cfqq; @@ -3582,9 +3469,16 @@ cfq_set_request(struct request_queue *q, struct request *rq, gfp_t gfp_mask) might_sleep_if(gfp_mask & __GFP_WAIT); spin_lock_irq(q->queue_lock); - cic = cfq_get_cic(cfqd, gfp_mask); - if (!cic) - goto queue_fail; + + /* handle changed notifications */ + if (unlikely(cic->icq.changed)) { + if (test_and_clear_bit(ICQ_IOPRIO_CHANGED, &cic->icq.changed)) + changed_ioprio(cic); +#ifdef CONFIG_CFQ_GROUP_IOSCHED + if (test_and_clear_bit(ICQ_CGROUP_CHANGED, &cic->icq.changed)) + changed_cgroup(cic); +#endif + } new_queue: cfqq = cic_to_cfqq(cic, is_sync); @@ -3615,17 +3509,10 @@ new_queue: cfqq->allocated[rw]++; cfqq->ref++; - rq->elv.icq = &cic->icq; rq->elv.priv[0] = cfqq; rq->elv.priv[1] = cfq_ref_get_cfqg(cfqq->cfqg); spin_unlock_irq(q->queue_lock); return 0; - -queue_fail: - cfq_schedule_dispatch(cfqd); - spin_unlock_irq(q->queue_lock); - cfq_log(cfqd, "set_request fail"); - return 1; } static void cfq_kick_queue(struct work_struct *work) |