--- /dev/null
+CFQ ioscheduler tunables
+========================
+
+slice_idle
+----------
+This specifies how long CFQ should idle for next request on certain cfq queues
+(for sequential workloads) and service trees (for random workloads) before
+queue is expired and CFQ selects next queue to dispatch from.
+
+By default slice_idle is a non-zero value. That means by default we idle on
+queues/service trees. This can be very helpful on highly seeky media like
+single spindle SATA/SAS disks where we can cut down on overall number of
+seeks and see improved throughput.
+
+Setting slice_idle to 0 will remove all the idling on queues/service tree
+level and one should see an overall improved throughput on faster storage
+devices like multiple SATA/SAS disks in hardware RAID configuration. The down
+side is that isolation provided from WRITES also goes down and notion of
+IO priority becomes weaker.
+
+So depending on storage and workload, it might be useful to set slice_idle=0.
+In general I think for SATA/SAS disks and software RAID of SATA/SAS disks
+keeping slice_idle enabled should be useful. For any configurations where
+there are multiple spindles behind single LUN (Host based hardware RAID
+controller or for storage arrays), setting slice_idle=0 might end up in better
+throughput and acceptable latencies.
+
+CFQ IOPS Mode for group scheduling
+===================================
+Basic CFQ design is to provide priority based time slices. Higher priority
+process gets bigger time slice and lower priority process gets smaller time
+slice. Measuring time becomes harder if storage is fast and supports NCQ and
+it would be better to dispatch multiple requests from multiple cfq queues in
+request queue at a time. In such scenario, it is not possible to measure time
+consumed by single queue accurately.
+
+What is possible though is to measure number of requests dispatched from a
+single queue and also allow dispatch from multiple cfq queue at the same time.
+This effectively becomes the fairness in terms of IOPS (IO operations per
+second).
+
+If one sets slice_idle=0 and if storage supports NCQ, CFQ internally switches
+to IOPS mode and starts providing fairness in terms of number of requests
+dispatched. Note that this mode switching takes effect only for group
+scheduling. For non-cgroup users nothing should change.
CFQ sysfs tunable
=================
/sys/block/<disk>/queue/iosched/group_isolation
+-----------------------------------------------
If group_isolation=1, it provides stronger isolation between groups at the
expense of throughput. By default group_isolation is 0. In general that
and one wants stronger isolation between groups, then set group_isolation=1
but this will come at cost of reduced throughput.
+/sys/block/<disk>/queue/iosched/slice_idle
+------------------------------------------
+On a faster hardware CFQ can be slow, especially with sequential workload.
+This happens because CFQ idles on a single queue and single queue might not
+drive deeper request queue depths to keep the storage busy. In such scenarios
+one can try setting slice_idle=0 and that would switch CFQ to IOPS
+(IO operations per second) mode on NCQ supporting hardware.
+
+That means CFQ will not idle between cfq queues of a cfq group and hence be
+able to driver higher queue depth and achieve better throughput. That also
+means that cfq provides fairness among groups in terms of IOPS and not in
+terms of disk time.
+
+/sys/block/<disk>/queue/iosched/group_idle
+------------------------------------------
+If one disables idling on individual cfq queues and cfq service trees by
+setting slice_idle=0, group_idle kicks in. That means CFQ will still idle
+on the group in an attempt to provide fairness among groups.
+
+By default group_idle is same as slice_idle and does not do anything if
+slice_idle is enabled.
+
+One can experience an overall throughput drop if you have created multiple
+groups and put applications in that group which are not driving enough
+IO to keep disk busy. In that case set group_idle=0, and CFQ will not idle
+on individual groups and throughput should improve.
+
What works
==========
- Currently only sync IO queues are support. All the buffered writes are
/* Currently we do not support hierarchy deeper than two level (0,1) */
if (parent != cgroup->top_cgroup)
- return ERR_PTR(-EINVAL);
+ return ERR_PTR(-EPERM);
blkcg = kzalloc(sizeof(*blkcg), GFP_KERNEL);
if (!blkcg)
int el_ret;
unsigned int bytes = bio->bi_size;
const unsigned short prio = bio_prio(bio);
- const bool sync = (bio->bi_rw & REQ_SYNC);
- const bool unplug = (bio->bi_rw & REQ_UNPLUG);
- const unsigned int ff = bio->bi_rw & REQ_FAILFAST_MASK;
+ const bool sync = !!(bio->bi_rw & REQ_SYNC);
+ const bool unplug = !!(bio->bi_rw & REQ_UNPLUG);
+ const unsigned long ff = bio->bi_rw & REQ_FAILFAST_MASK;
int rw_flags;
if ((bio->bi_rw & REQ_HARDBARRIER) &&
kobject_uevent(&q->kobj, KOBJ_REMOVE);
kobject_del(&q->kobj);
blk_trace_remove_sysfs(disk_to_dev(disk));
+ kobject_put(&dev->kobj);
return ret;
}
static inline int blk_cpu_to_group(int cpu)
{
+ int group = NR_CPUS;
#ifdef CONFIG_SCHED_MC
const struct cpumask *mask = cpu_coregroup_mask(cpu);
- return cpumask_first(mask);
+ group = cpumask_first(mask);
#elif defined(CONFIG_SCHED_SMT)
- return cpumask_first(topology_thread_cpumask(cpu));
+ group = cpumask_first(topology_thread_cpumask(cpu));
#else
return cpu;
#endif
+ if (likely(group < NR_CPUS))
+ return group;
+ return cpu;
}
/*
static int cfq_slice_async = HZ / 25;
static const int cfq_slice_async_rq = 2;
static int cfq_slice_idle = HZ / 125;
+static int cfq_group_idle = HZ / 125;
static const int cfq_target_latency = HZ * 3/10; /* 300 ms */
static const int cfq_hist_divisor = 4;
struct cfq_queue *new_cfqq;
struct cfq_group *cfqg;
struct cfq_group *orig_cfqg;
+ /* Number of sectors dispatched from queue in single dispatch round */
+ unsigned long nr_sectors;
};
/*
struct hlist_node cfqd_node;
atomic_t ref;
#endif
+ /* number of requests that are on the dispatch list or inside driver */
+ int dispatched;
};
/*
unsigned int cfq_slice[2];
unsigned int cfq_slice_async_rq;
unsigned int cfq_slice_idle;
+ unsigned int cfq_group_idle;
unsigned int cfq_latency;
unsigned int cfq_group_isolation;
&cfqg->service_trees[i][j]: NULL) \
+static inline bool iops_mode(struct cfq_data *cfqd)
+{
+ /*
+ * If we are not idling on queues and it is a NCQ drive, parallel
+ * execution of requests is on and measuring time is not possible
+ * in most of the cases until and unless we drive shallower queue
+ * depths and that becomes a performance bottleneck. In such cases
+ * switch to start providing fairness in terms of number of IOs.
+ */
+ if (!cfqd->cfq_slice_idle && cfqd->hw_tag)
+ return true;
+ else
+ return false;
+}
+
static inline enum wl_prio_t cfqq_prio(struct cfq_queue *cfqq)
{
if (cfq_class_idle(cfqq))
slice_used = cfqq->allocated_slice;
}
- cfq_log_cfqq(cfqq->cfqd, cfqq, "sl_used=%u", slice_used);
return slice_used;
}
struct cfq_queue *cfqq)
{
struct cfq_rb_root *st = &cfqd->grp_service_tree;
- unsigned int used_sl, charge_sl;
+ unsigned int used_sl, charge;
int nr_sync = cfqg->nr_cfqq - cfqg_busy_async_queues(cfqd, cfqg)
- cfqg->service_tree_idle.count;
BUG_ON(nr_sync < 0);
- used_sl = charge_sl = cfq_cfqq_slice_usage(cfqq);
+ used_sl = charge = cfq_cfqq_slice_usage(cfqq);
- if (!cfq_cfqq_sync(cfqq) && !nr_sync)
- charge_sl = cfqq->allocated_slice;
+ if (iops_mode(cfqd))
+ charge = cfqq->slice_dispatch;
+ else if (!cfq_cfqq_sync(cfqq) && !nr_sync)
+ charge = cfqq->allocated_slice;
/* Can't update vdisktime while group is on service tree */
cfq_rb_erase(&cfqg->rb_node, st);
- cfqg->vdisktime += cfq_scale_slice(charge_sl, cfqg);
+ cfqg->vdisktime += cfq_scale_slice(charge, cfqg);
__cfq_group_service_tree_add(st, cfqg);
/* This group is being expired. Save the context */
cfq_log_cfqg(cfqd, cfqg, "served: vt=%llu min_vt=%llu", cfqg->vdisktime,
st->min_vdisktime);
+ cfq_log_cfqq(cfqq->cfqd, cfqq, "sl_used=%u disp=%u charge=%u iops=%u"
+ " sect=%u", used_sl, cfqq->slice_dispatch, charge,
+ iops_mode(cfqd), cfqq->nr_sectors);
cfq_blkiocg_update_timeslice_used(&cfqg->blkg, used_sl);
cfq_blkiocg_set_start_empty_time(&cfqg->blkg);
}
cfqq->allocated_slice = 0;
cfqq->slice_end = 0;
cfqq->slice_dispatch = 0;
+ cfqq->nr_sectors = 0;
cfq_clear_cfqq_wait_request(cfqq);
cfq_clear_cfqq_must_dispatch(cfqq);
BUG_ON(!service_tree);
BUG_ON(!service_tree->count);
+ if (!cfqd->cfq_slice_idle)
+ return false;
+
/* We never do for idle class queues. */
if (prio == IDLE_WORKLOAD)
return false;
{
struct cfq_queue *cfqq = cfqd->active_queue;
struct cfq_io_context *cic;
- unsigned long sl;
+ unsigned long sl, group_idle = 0;
/*
* SSD device without seek penalty, disable idling. But only do so
/*
* idle is disabled, either manually or by past process history
*/
- if (!cfqd->cfq_slice_idle || !cfq_should_idle(cfqd, cfqq))
- return;
+ if (!cfq_should_idle(cfqd, cfqq)) {
+ /* no queue idling. Check for group idling */
+ if (cfqd->cfq_group_idle)
+ group_idle = cfqd->cfq_group_idle;
+ else
+ return;
+ }
/*
* still active requests from this queue, don't idle
return;
}
+ /* There are other queues in the group, don't do group idle */
+ if (group_idle && cfqq->cfqg->nr_cfqq > 1)
+ return;
+
cfq_mark_cfqq_wait_request(cfqq);
- sl = cfqd->cfq_slice_idle;
+ if (group_idle)
+ sl = cfqd->cfq_group_idle;
+ else
+ sl = cfqd->cfq_slice_idle;
mod_timer(&cfqd->idle_slice_timer, jiffies + sl);
cfq_blkiocg_update_set_idle_time_stats(&cfqq->cfqg->blkg);
- cfq_log_cfqq(cfqd, cfqq, "arm_idle: %lu", sl);
+ cfq_log_cfqq(cfqd, cfqq, "arm_idle: %lu group_idle: %d", sl,
+ group_idle ? 1 : 0);
}
/*
cfqq->next_rq = cfq_find_next_rq(cfqd, cfqq, rq);
cfq_remove_request(rq);
cfqq->dispatched++;
+ (RQ_CFQG(rq))->dispatched++;
elv_dispatch_sort(q, rq);
cfqd->rq_in_flight[cfq_cfqq_sync(cfqq)]++;
+ cfqq->nr_sectors += blk_rq_sectors(rq);
cfq_blkiocg_update_dispatch_stats(&cfqq->cfqg->blkg, blk_rq_bytes(rq),
rq_data_dir(rq), rq_is_sync(rq));
}
cfqq = NULL;
goto keep_queue;
} else
- goto expire;
+ goto check_group_idle;
}
/*
* flight or is idling for a new request, allow either of these
* conditions to happen (or time out) before selecting a new queue.
*/
- if (timer_pending(&cfqd->idle_slice_timer) ||
- (cfqq->dispatched && cfq_should_idle(cfqd, cfqq))) {
+ if (timer_pending(&cfqd->idle_slice_timer)) {
+ cfqq = NULL;
+ goto keep_queue;
+ }
+
+ if (cfqq->dispatched && cfq_should_idle(cfqd, cfqq)) {
+ cfqq = NULL;
+ goto keep_queue;
+ }
+
+ /*
+ * If group idle is enabled and there are requests dispatched from
+ * this group, wait for requests to complete.
+ */
+check_group_idle:
+ if (cfqd->cfq_group_idle && cfqq->cfqg->nr_cfqq == 1
+ && cfqq->cfqg->dispatched) {
cfqq = NULL;
goto keep_queue;
}
WARN_ON(!cfqq->dispatched);
cfqd->rq_in_driver--;
cfqq->dispatched--;
+ (RQ_CFQG(rq))->dispatched--;
cfq_blkiocg_update_completion_stats(&cfqq->cfqg->blkg,
rq_start_time_ns(rq), rq_io_start_time_ns(rq),
rq_data_dir(rq), rq_is_sync(rq));
* the queue.
*/
if (cfq_should_wait_busy(cfqd, cfqq)) {
- cfqq->slice_end = jiffies + cfqd->cfq_slice_idle;
+ unsigned long extend_sl = cfqd->cfq_slice_idle;
+ if (!cfqd->cfq_slice_idle)
+ extend_sl = cfqd->cfq_group_idle;
+ cfqq->slice_end = jiffies + extend_sl;
cfq_mark_cfqq_wait_busy(cfqq);
cfq_log_cfqq(cfqd, cfqq, "will busy wait");
}
cfqd->cfq_slice[1] = cfq_slice_sync;
cfqd->cfq_slice_async_rq = cfq_slice_async_rq;
cfqd->cfq_slice_idle = cfq_slice_idle;
+ cfqd->cfq_group_idle = cfq_group_idle;
cfqd->cfq_latency = 1;
cfqd->cfq_group_isolation = 0;
cfqd->hw_tag = -1;
SHOW_FUNCTION(cfq_back_seek_max_show, cfqd->cfq_back_max, 0);
SHOW_FUNCTION(cfq_back_seek_penalty_show, cfqd->cfq_back_penalty, 0);
SHOW_FUNCTION(cfq_slice_idle_show, cfqd->cfq_slice_idle, 1);
+SHOW_FUNCTION(cfq_group_idle_show, cfqd->cfq_group_idle, 1);
SHOW_FUNCTION(cfq_slice_sync_show, cfqd->cfq_slice[1], 1);
SHOW_FUNCTION(cfq_slice_async_show, cfqd->cfq_slice[0], 1);
SHOW_FUNCTION(cfq_slice_async_rq_show, cfqd->cfq_slice_async_rq, 0);
STORE_FUNCTION(cfq_back_seek_penalty_store, &cfqd->cfq_back_penalty, 1,
UINT_MAX, 0);
STORE_FUNCTION(cfq_slice_idle_store, &cfqd->cfq_slice_idle, 0, UINT_MAX, 1);
+STORE_FUNCTION(cfq_group_idle_store, &cfqd->cfq_group_idle, 0, UINT_MAX, 1);
STORE_FUNCTION(cfq_slice_sync_store, &cfqd->cfq_slice[1], 1, UINT_MAX, 1);
STORE_FUNCTION(cfq_slice_async_store, &cfqd->cfq_slice[0], 1, UINT_MAX, 1);
STORE_FUNCTION(cfq_slice_async_rq_store, &cfqd->cfq_slice_async_rq, 1,
CFQ_ATTR(slice_async),
CFQ_ATTR(slice_async_rq),
CFQ_ATTR(slice_idle),
+ CFQ_ATTR(group_idle),
CFQ_ATTR(low_latency),
CFQ_ATTR(group_isolation),
__ATTR_NULL
if (!cfq_slice_idle)
cfq_slice_idle = 1;
+#ifdef CONFIG_CFQ_GROUP_IOSCHED
+ if (!cfq_group_idle)
+ cfq_group_idle = 1;
+#else
+ cfq_group_idle = 0;
+#endif
if (cfq_slab_setup())
return -ENOMEM;
{
struct elevator_queue *old_elevator, *e;
void *data;
+ int err;
/*
* Allocate new elevator
*/
e = elevator_alloc(q, new_e);
if (!e)
- return 0;
+ return -ENOMEM;
data = elevator_init_queue(q, e);
if (!data) {
kobject_put(&e->kobj);
- return 0;
+ return -ENOMEM;
}
/*
__elv_unregister_queue(old_elevator);
- if (elv_register_queue(q))
+ err = elv_register_queue(q);
+ if (err)
goto fail_register;
/*
blk_add_trace_msg(q, "elv switch: %s", e->elevator_type->elevator_name);
- return 1;
+ return 0;
fail_register:
/*
queue_flag_clear(QUEUE_FLAG_ELVSWITCH, q);
spin_unlock_irq(q->queue_lock);
- return 0;
+ return err;
}
-ssize_t elv_iosched_store(struct request_queue *q, const char *name,
- size_t count)
+/*
+ * Switch this queue to the given IO scheduler.
+ */
+int elevator_change(struct request_queue *q, const char *name)
{
char elevator_name[ELV_NAME_MAX];
struct elevator_type *e;
if (!q->elevator)
- return count;
+ return -ENXIO;
strlcpy(elevator_name, name, sizeof(elevator_name));
e = elevator_get(strstrip(elevator_name));
if (!strcmp(elevator_name, q->elevator->elevator_type->elevator_name)) {
elevator_put(e);
- return count;
+ return 0;
}
- if (!elevator_switch(q, e))
- printk(KERN_ERR "elevator: switch to %s failed\n",
- elevator_name);
- return count;
+ return elevator_switch(q, e);
+}
+EXPORT_SYMBOL(elevator_change);
+
+ssize_t elv_iosched_store(struct request_queue *q, const char *name,
+ size_t count)
+{
+ int ret;
+
+ if (!q->elevator)
+ return count;
+
+ ret = elevator_change(q, name);
+ if (!ret)
+ return count;
+
+ printk(KERN_ERR "elevator: switch to %s failed\n", name);
+ return ret;
}
ssize_t elv_iosched_show(struct request_queue *q, char *name)
spin_lock_irqsave(&h->lock, flags);
addQ(&h->reqQ, c);
h->Qdepth++;
+ if (h->Qdepth > h->maxQsinceinit)
+ h->maxQsinceinit = h->Qdepth;
start_io(h);
spin_unlock_irqrestore(&h->lock, flags);
}
misc_fw_support = readl(&cfgtable->misc_fw_support);
use_doorbell = misc_fw_support & MISC_FW_DOORBELL_RESET;
+ /* The doorbell reset seems to cause lockups on some Smart
+ * Arrays (e.g. P410, P410i, maybe others). Until this is
+ * fixed or at least isolated, avoid the doorbell reset.
+ */
+ use_doorbell = 0;
+
rc = cciss_controller_hard_reset(pdev, vaddr, use_doorbell);
if (rc)
goto unmap_cfgtable;
h->scatter_list = kmalloc(h->max_commands *
sizeof(struct scatterlist *),
GFP_KERNEL);
+ if (!h->scatter_list)
+ goto clean4;
+
for (k = 0; k < h->nr_cmds; k++) {
h->scatter_list[k] = kmalloc(sizeof(struct scatterlist) *
h->maxsgentries,
pos = ((loff_t) bio->bi_sector << 9) + lo->lo_offset;
if (bio_rw(bio) == WRITE) {
- bool barrier = (bio->bi_rw & REQ_HARDBARRIER);
+ bool barrier = !!(bio->bi_rw & REQ_HARDBARRIER);
struct file *file = lo->lo_backing_file;
if (barrier) {
host->breq->queuedata = host;
/* mflash is random device, thanx for the noop */
- elevator_exit(host->breq->elevator);
- err = elevator_init(host->breq, "noop");
+ err = elevator_change(host->breq, "noop");
if (err) {
printk(KERN_ERR "%s:%d (elevator_init) fail\n",
__func__, __LINE__);
if (!blkdat->request_queue)
return -ENOMEM;
- elevator_exit(blkdat->request_queue->elevator);
- rc = elevator_init(blkdat->request_queue, "noop");
+ rc = elevator_change(blkdat->request_queue, "noop");
if (rc)
goto cleanup_queue;
/* Allocate kernel buffer for protection data */
len = sectors * blk_integrity_tuple_size(bi);
- buf = kmalloc(len, GFP_NOIO | __GFP_NOFAIL | q->bounce_gfp);
+ buf = kmalloc(len, GFP_NOIO | q->bounce_gfp);
if (unlikely(buf == NULL)) {
printk(KERN_ERR "could not allocate integrity buffer\n");
- return -EIO;
+ return -ENOMEM;
}
end = (((unsigned long) buf) + len + PAGE_SIZE - 1) >> PAGE_SHIFT;
wb->last_active = jiffies;
set_current_state(TASK_INTERRUPTIBLE);
- if (!list_empty(&bdi->work_list)) {
+ if (!list_empty(&bdi->work_list) || kthread_should_stop()) {
__set_current_state(TASK_RUNNING);
continue;
}
extern int elevator_init(struct request_queue *, char *);
extern void elevator_exit(struct elevator_queue *);
+extern int elevator_change(struct request_queue *, const char *);
extern int elv_rq_merge_ok(struct request *, struct bio *);
/*
left -= sg_size;
sg = alloc_fn(alloc_size, gfp_mask);
- if (unlikely(!sg))
- return -ENOMEM;
+ if (unlikely(!sg)) {
+ /*
+ * Adjust entry count to reflect that the last
+ * entry of the previous table won't be used for
+ * linkage. Without this, sg_kfree() may get
+ * confused.
+ */
+ if (prv)
+ table->nents = ++table->orig_nents;
+
+ return -ENOMEM;
+ }
sg_init_table(sg, alloc_size);
table->nents = table->orig_nents += sg_size;
switch (action) {
case FORK_THREAD:
__set_current_state(TASK_RUNNING);
- task = kthread_run(bdi_writeback_thread, &bdi->wb, "flush-%s",
- dev_name(bdi->dev));
+ task = kthread_create(bdi_writeback_thread, &bdi->wb,
+ "flush-%s", dev_name(bdi->dev));
if (IS_ERR(task)) {
/*
* If thread creation fails, force writeout of
/*
* The spinlock makes sure we do not lose
* wake-ups when racing with 'bdi_queue_work()'.
+ * And as soon as the bdi thread is visible, we
+ * can start it.
*/
spin_lock_bh(&bdi->wb_lock);
bdi->wb.task = task;
spin_unlock_bh(&bdi->wb_lock);
+ wake_up_process(task);
}
break;
projects / mv-sheeva.git / commitdiff