| Message ID | 20240410101858.1149134-1-dongliang.cui@unisoc.com (mailing list archive) |
|---|---|
| State | New, archived |
| Headers | show |
| Series | block: add max_dispatch to sysfs | expand |
On 4/10/24 4:18 AM, Dongliang Cui wrote: > The default configuration in the current code is that when the device > is not busy, a single dispatch will attempt to pull 'nr_requests' > requests out of the schedule queue. > > I tried to track the dispatch process: > > COMM TYPE SEC_START IOPRIO INDEX > fio-17304 R 196798040 0x2005 0 > fio-17306 R 197060504 0x2005 1 > fio-17307 R 197346904 0x2005 2 > fio-17308 R 197609400 0x2005 3 > fio-17309 R 197873048 0x2005 4 > fio-17310 R 198134936 0x2005 5 > ... > fio-17237 R 197122936 0x0 57 > fio-17238 R 197384984 0x0 58 > <...>-17239 R 197647128 0x0 59 > fio-17240 R 197909208 0x0 60 > fio-17241 R 198171320 0x0 61 > fio-17242 R 198433432 0x0 62 > fio-17300 R 195744088 0x2005 0 > fio-17301 R 196008504 0x2005 0 > > The above data is calculated based on the block event trace, with each > column containing: process name, request type, sector start address, > IO priority. > > The INDEX represents the order in which the requests are extracted from > the scheduler queue during a single dispatch process. > > Some low-speed devices cannot process these requests at once, and they will > be requeued to hctx->dispatch and wait for the next issuance. > > There will be a problem here, when the IO priority is enabled, if you try > to dispatch "nr_request" requests at once, the IO priority will be ignored > from the scheduler queue and all requests will be extracted. > > In this scenario, if a high priority request is inserted into the scheduler > queue, it needs to wait for the low priority request in the hctx->dispatch > to be processed first. > > --------------------dispatch 1st---------------------- > fio-17241 R 198171320 0x0 61 > fio-17242 R 198433432 0x0 62 > --------------------dispatch 2nd---------------------- > fio-17300 R 195744088 0x2005 0 > > In certain scenarios, we hope that requests can be processed in order of io > priority as much as possible. > > Maybe max_dispatch should not be a fixed value, but can be adjusted > according to device conditions. > > So we give a interface to control the maximum value of single dispatch > so that users can configure it according to devices characteristics. I agree that pulling 'nr_requests' out of the scheduler will kind of defeat the purpose of the scheduler to some extent. But rather than add another knob that nobody knows about or ever will touch (and extra queue variables that just take up space), why not just default to something a bit saner? Eg we could default to 1/8 or 1/4 of the scheduler depth instead.
On 4/10/24 22:17, Jens Axboe wrote: > On 4/10/24 4:18 AM, Dongliang Cui wrote: >> The default configuration in the current code is that when the device >> is not busy, a single dispatch will attempt to pull 'nr_requests' >> requests out of the schedule queue. >> >> I tried to track the dispatch process: >> >> COMM TYPE SEC_START IOPRIO INDEX >> fio-17304 R 196798040 0x2005 0 >> fio-17306 R 197060504 0x2005 1 >> fio-17307 R 197346904 0x2005 2 >> fio-17308 R 197609400 0x2005 3 >> fio-17309 R 197873048 0x2005 4 >> fio-17310 R 198134936 0x2005 5 >> ... >> fio-17237 R 197122936 0x0 57 >> fio-17238 R 197384984 0x0 58 >> <...>-17239 R 197647128 0x0 59 >> fio-17240 R 197909208 0x0 60 >> fio-17241 R 198171320 0x0 61 >> fio-17242 R 198433432 0x0 62 >> fio-17300 R 195744088 0x2005 0 >> fio-17301 R 196008504 0x2005 0 >> >> The above data is calculated based on the block event trace, with each >> column containing: process name, request type, sector start address, >> IO priority. >> >> The INDEX represents the order in which the requests are extracted from >> the scheduler queue during a single dispatch process. >> >> Some low-speed devices cannot process these requests at once, and they will >> be requeued to hctx->dispatch and wait for the next issuance. >> >> There will be a problem here, when the IO priority is enabled, if you try >> to dispatch "nr_request" requests at once, the IO priority will be ignored >> from the scheduler queue and all requests will be extracted. >> >> In this scenario, if a high priority request is inserted into the scheduler >> queue, it needs to wait for the low priority request in the hctx->dispatch >> to be processed first. >> >> --------------------dispatch 1st---------------------- >> fio-17241 R 198171320 0x0 61 >> fio-17242 R 198433432 0x0 62 >> --------------------dispatch 2nd---------------------- >> fio-17300 R 195744088 0x2005 0 >> >> In certain scenarios, we hope that requests can be processed in order of io >> priority as much as possible. >> >> Maybe max_dispatch should not be a fixed value, but can be adjusted >> according to device conditions. >> >> So we give a interface to control the maximum value of single dispatch >> so that users can configure it according to devices characteristics. > > I agree that pulling 'nr_requests' out of the scheduler will kind of > defeat the purpose of the scheduler to some extent. But rather than add > another knob that nobody knows about or ever will touch (and extra queue > variables that just take up space), why not just default to something a > bit saner? Eg we could default to 1/8 or 1/4 of the scheduler depth > instead. Why not default to pulling what can actually be executed, that is, up to the number of free hw tags / budget ? Anything more than that will be requeued anyway.
On Thu, Apr 11, 2024 at 1:19 PM Damien Le Moal <dlemoal@kernel.org> wrote: > > On 4/10/24 22:17, Jens Axboe wrote: > > On 4/10/24 4:18 AM, Dongliang Cui wrote: > >> The default configuration in the current code is that when the device > >> is not busy, a single dispatch will attempt to pull 'nr_requests' > >> requests out of the schedule queue. > >> > >> I tried to track the dispatch process: > >> > >> COMM TYPE SEC_START IOPRIO INDEX > >> fio-17304 R 196798040 0x2005 0 > >> fio-17306 R 197060504 0x2005 1 > >> fio-17307 R 197346904 0x2005 2 > >> fio-17308 R 197609400 0x2005 3 > >> fio-17309 R 197873048 0x2005 4 > >> fio-17310 R 198134936 0x2005 5 > >> ... > >> fio-17237 R 197122936 0x0 57 > >> fio-17238 R 197384984 0x0 58 > >> <...>-17239 R 197647128 0x0 59 > >> fio-17240 R 197909208 0x0 60 > >> fio-17241 R 198171320 0x0 61 > >> fio-17242 R 198433432 0x0 62 > >> fio-17300 R 195744088 0x2005 0 > >> fio-17301 R 196008504 0x2005 0 > >> > >> The above data is calculated based on the block event trace, with each > >> column containing: process name, request type, sector start address, > >> IO priority. > >> > >> The INDEX represents the order in which the requests are extracted from > >> the scheduler queue during a single dispatch process. > >> > >> Some low-speed devices cannot process these requests at once, and they will > >> be requeued to hctx->dispatch and wait for the next issuance. > >> > >> There will be a problem here, when the IO priority is enabled, if you try > >> to dispatch "nr_request" requests at once, the IO priority will be ignored > >> from the scheduler queue and all requests will be extracted. > >> > >> In this scenario, if a high priority request is inserted into the scheduler > >> queue, it needs to wait for the low priority request in the hctx->dispatch > >> to be processed first. > >> > >> --------------------dispatch 1st---------------------- > >> fio-17241 R 198171320 0x0 61 > >> fio-17242 R 198433432 0x0 62 > >> --------------------dispatch 2nd---------------------- > >> fio-17300 R 195744088 0x2005 0 > >> > >> In certain scenarios, we hope that requests can be processed in order of io > >> priority as much as possible. > >> > >> Maybe max_dispatch should not be a fixed value, but can be adjusted > >> according to device conditions. > >> > >> So we give a interface to control the maximum value of single dispatch > >> so that users can configure it according to devices characteristics. > > > > I agree that pulling 'nr_requests' out of the scheduler will kind of > > defeat the purpose of the scheduler to some extent. But rather than add > > another knob that nobody knows about or ever will touch (and extra queue > > variables that just take up space), why not just default to something a > > bit saner? Eg we could default to 1/8 or 1/4 of the scheduler depth > > instead. Current mechanism will try to pulling everything out of the scheduler, regardless of the priority of request. Perhaps reducing the queue depth does not solve the priority disorder scenario. Reducing depth may also weakens the role of priorify. > > Why not default to pulling what can actually be executed, that is, up to the > number of free hw tags / budget ? Anything more than that will be requeued anyway. The process of pulling the request out of schedule will try to obtain the hw tags. If hw tags are obtained, request will continue to be pulled out of scheduler. However, for slow devices, the number of hw tags is generally greater than the number of requests that the device can currently handle, and unprocessed requests may not be arranged in dispatch_list in order of priority. And for budget, I found that the budget is set by the driver. Some slow devices, such as emmc, do not register this interface, unable to set budget. > -- > Damien Le Moal > Western Digital Research >
On 4/12/24 16:25, dongliang cui wrote: > On Thu, Apr 11, 2024 at 1:19 PM Damien Le Moal <dlemoal@kernel.org> wrote: >> >> On 4/10/24 22:17, Jens Axboe wrote: >>> On 4/10/24 4:18 AM, Dongliang Cui wrote: >>>> The default configuration in the current code is that when the device >>>> is not busy, a single dispatch will attempt to pull 'nr_requests' >>>> requests out of the schedule queue. >>>> >>>> I tried to track the dispatch process: >>>> >>>> COMM TYPE SEC_START IOPRIO INDEX >>>> fio-17304 R 196798040 0x2005 0 >>>> fio-17306 R 197060504 0x2005 1 >>>> fio-17307 R 197346904 0x2005 2 >>>> fio-17308 R 197609400 0x2005 3 >>>> fio-17309 R 197873048 0x2005 4 >>>> fio-17310 R 198134936 0x2005 5 >>>> ... >>>> fio-17237 R 197122936 0x0 57 >>>> fio-17238 R 197384984 0x0 58 >>>> <...>-17239 R 197647128 0x0 59 >>>> fio-17240 R 197909208 0x0 60 >>>> fio-17241 R 198171320 0x0 61 >>>> fio-17242 R 198433432 0x0 62 >>>> fio-17300 R 195744088 0x2005 0 >>>> fio-17301 R 196008504 0x2005 0 >>>> >>>> The above data is calculated based on the block event trace, with each >>>> column containing: process name, request type, sector start address, >>>> IO priority. >>>> >>>> The INDEX represents the order in which the requests are extracted from >>>> the scheduler queue during a single dispatch process. >>>> >>>> Some low-speed devices cannot process these requests at once, and they will >>>> be requeued to hctx->dispatch and wait for the next issuance. >>>> >>>> There will be a problem here, when the IO priority is enabled, if you try >>>> to dispatch "nr_request" requests at once, the IO priority will be ignored >>>> from the scheduler queue and all requests will be extracted. >>>> >>>> In this scenario, if a high priority request is inserted into the scheduler >>>> queue, it needs to wait for the low priority request in the hctx->dispatch >>>> to be processed first. >>>> >>>> --------------------dispatch 1st---------------------- >>>> fio-17241 R 198171320 0x0 61 >>>> fio-17242 R 198433432 0x0 62 >>>> --------------------dispatch 2nd---------------------- >>>> fio-17300 R 195744088 0x2005 0 >>>> >>>> In certain scenarios, we hope that requests can be processed in order of io >>>> priority as much as possible. >>>> >>>> Maybe max_dispatch should not be a fixed value, but can be adjusted >>>> according to device conditions. >>>> >>>> So we give a interface to control the maximum value of single dispatch >>>> so that users can configure it according to devices characteristics. >>> >>> I agree that pulling 'nr_requests' out of the scheduler will kind of >>> defeat the purpose of the scheduler to some extent. But rather than add >>> another knob that nobody knows about or ever will touch (and extra queue >>> variables that just take up space), why not just default to something a >>> bit saner? Eg we could default to 1/8 or 1/4 of the scheduler depth >>> instead. > Current mechanism will try to pulling everything out of the scheduler, > regardless > of the priority of request. Perhaps reducing the queue depth does not solve the > priority disorder scenario. Reducing depth may also weakens the role > of priorify. > >> >> Why not default to pulling what can actually be executed, that is, up to the >> number of free hw tags / budget ? Anything more than that will be requeued anyway. > The process of pulling the request out of schedule will try to obtain > the hw tags. > If hw tags are obtained, request will continue to be pulled out of scheduler. > However, for slow devices, the number of hw tags is generally greater than the > number of requests that the device can currently handle, and > unprocessed requests > may not be arranged in dispatch_list in order of priority. What ? If the number of hw tags you have for your device is larger than what the device can actually handle, that means that your device driver is very buggy. A hw tag is a direct identifier of queued commands on the device side. So you cannot have more than what the device allows. May be you are confusing this with the fact that even if you have the correct maximum number of hw tags, the device may sometimes refuse to accept a new command using a free tag. E.g. a scsi device may return TSF (task set full) even if its queue is not full (some hw tags are free). But that is not the common case. As for the commands not being arranged in priority order when submitted, that is also strange. You are going to get the commands from the scheduler in priority order, so they will be sorted naturally as such in the dispatch queue, unless you get a lot of requeue from the device (again, generally not the usual case). > And for budget, I found that the budget is set by the driver. Some > slow devices, > such as emmc, do not register this interface, unable to set budget. May be that is exactly what you need to do to solve your issue: implement that for emmc. Have you tried ?
diff --git a/block/blk-core.c b/block/blk-core.c index de771093b526..f5a917085eae 100644 --- a/block/blk-core.c +++ b/block/blk-core.c @@ -442,6 +442,7 @@ struct request_queue *blk_alloc_queue(int node_id) blk_set_default_limits(&q->limits); q->nr_requests = BLKDEV_DEFAULT_RQ; + q->max_dispatch = BLKDEV_DEFAULT_RQ; return q; diff --git a/block/blk-mq-sched.c b/block/blk-mq-sched.c index 451a2c1f1f32..019958c0a4c3 100644 --- a/block/blk-mq-sched.c +++ b/block/blk-mq-sched.c @@ -97,7 +97,7 @@ static int __blk_mq_do_dispatch_sched(struct blk_mq_hw_ctx *hctx) if (hctx->dispatch_busy) max_dispatch = 1; else - max_dispatch = hctx->queue->nr_requests; + max_dispatch = hctx->queue->max_dispatch; do { struct request *rq; @@ -454,6 +454,8 @@ int blk_mq_init_sched(struct request_queue *q, struct elevator_type *e) q->nr_requests = 2 * min_t(unsigned int, q->tag_set->queue_depth, BLKDEV_DEFAULT_RQ); + q->max_dispatch = q->nr_requests; + if (blk_mq_is_shared_tags(flags)) { ret = blk_mq_init_sched_shared_tags(q); if (ret) diff --git a/block/blk-mq.c b/block/blk-mq.c index 2dc01551e27c..9c286001f429 100644 --- a/block/blk-mq.c +++ b/block/blk-mq.c @@ -4285,6 +4285,7 @@ int blk_mq_init_allocated_queue(struct blk_mq_tag_set *set, spin_lock_init(&q->requeue_lock); q->nr_requests = set->queue_depth; + q->max_dispatch = set->queue_depth; blk_mq_init_cpu_queues(q, set->nr_hw_queues); blk_mq_add_queue_tag_set(set, q); @@ -4634,6 +4635,8 @@ int blk_mq_update_nr_requests(struct request_queue *q, unsigned int nr) } if (!ret) { q->nr_requests = nr; + if (q->max_dispatch > nr) + q->max_dispatch = nr; if (blk_mq_is_shared_tags(set->flags)) { if (q->elevator) blk_mq_tag_update_sched_shared_tags(q); diff --git a/block/blk-sysfs.c b/block/blk-sysfs.c index 6b2429cad81a..909b5f158bd3 100644 --- a/block/blk-sysfs.c +++ b/block/blk-sysfs.c @@ -100,6 +100,36 @@ queue_ra_store(struct request_queue *q, const char *page, size_t count) return ret; } +static ssize_t queue_max_dispatch_show(struct request_queue *q, char *page) +{ + unsigned long max_dispatch; + + if (!q->disk) + return -EINVAL; + max_dispatch = q->max_dispatch; + return queue_var_show(max_dispatch, page); +} + +static ssize_t +queue_max_dispatch_store(struct request_queue *q, const char *page, size_t count) +{ + unsigned long max_dispatch; + ssize_t ret; + + if (!q->disk) + return -EINVAL; + + ret = queue_var_store(&max_dispatch, page, count); + if (ret < 0) + return ret; + + if (max_dispatch > q->nr_requests) + max_dispatch = q->nr_requests; + + q->max_dispatch = max_dispatch; + return ret; +} + static ssize_t queue_max_sectors_show(struct request_queue *q, char *page) { int max_sectors_kb = queue_max_sectors(q) >> 1; @@ -484,6 +514,7 @@ static struct queue_sysfs_entry _prefix##_entry = { \ QUEUE_RW_ENTRY(queue_requests, "nr_requests"); QUEUE_RW_ENTRY(queue_ra, "read_ahead_kb"); QUEUE_RW_ENTRY(queue_max_sectors, "max_sectors_kb"); +QUEUE_RW_ENTRY(queue_max_dispatch, "max_dispatch"); QUEUE_RO_ENTRY(queue_max_hw_sectors, "max_hw_sectors_kb"); QUEUE_RO_ENTRY(queue_max_segments, "max_segments"); QUEUE_RO_ENTRY(queue_max_integrity_segments, "max_integrity_segments"); @@ -614,6 +645,7 @@ QUEUE_RW_ENTRY(queue_wb_lat, "wbt_lat_usec"); static struct attribute *queue_attrs[] = { &queue_ra_entry.attr, &queue_max_hw_sectors_entry.attr, + &queue_max_dispatch_entry.attr, &queue_max_sectors_entry.attr, &queue_max_segments_entry.attr, &queue_max_discard_segments_entry.attr, diff --git a/include/linux/blkdev.h b/include/linux/blkdev.h index 99e4f5e72213..a96791b83977 100644 --- a/include/linux/blkdev.h +++ b/include/linux/blkdev.h @@ -434,6 +434,8 @@ struct request_queue { */ unsigned long nr_requests; /* Max # of requests */ + unsigned long max_dispatch; /* Max # of single dispatch */ + #ifdef CONFIG_BLK_INLINE_ENCRYPTION struct blk_crypto_profile *crypto_profile; struct kobject *crypto_kobject;
The default configuration in the current code is that when the device is not busy, a single dispatch will attempt to pull 'nr_requests' requests out of the schedule queue. I tried to track the dispatch process: COMM TYPE SEC_START IOPRIO INDEX fio-17304 R 196798040 0x2005 0 fio-17306 R 197060504 0x2005 1 fio-17307 R 197346904 0x2005 2 fio-17308 R 197609400 0x2005 3 fio-17309 R 197873048 0x2005 4 fio-17310 R 198134936 0x2005 5 ... fio-17237 R 197122936 0x0 57 fio-17238 R 197384984 0x0 58 <...>-17239 R 197647128 0x0 59 fio-17240 R 197909208 0x0 60 fio-17241 R 198171320 0x0 61 fio-17242 R 198433432 0x0 62 fio-17300 R 195744088 0x2005 0 fio-17301 R 196008504 0x2005 0 The above data is calculated based on the block event trace, with each column containing: process name, request type, sector start address, IO priority. The INDEX represents the order in which the requests are extracted from the scheduler queue during a single dispatch process. Some low-speed devices cannot process these requests at once, and they will be requeued to hctx->dispatch and wait for the next issuance. There will be a problem here, when the IO priority is enabled, if you try to dispatch "nr_request" requests at once, the IO priority will be ignored from the scheduler queue and all requests will be extracted. In this scenario, if a high priority request is inserted into the scheduler queue, it needs to wait for the low priority request in the hctx->dispatch to be processed first. --------------------dispatch 1st---------------------- fio-17241 R 198171320 0x0 61 fio-17242 R 198433432 0x0 62 --------------------dispatch 2nd---------------------- fio-17300 R 195744088 0x2005 0 In certain scenarios, we hope that requests can be processed in order of io priority as much as possible. Maybe max_dispatch should not be a fixed value, but can be adjusted according to device conditions. So we give a interface to control the maximum value of single dispatch so that users can configure it according to devices characteristics. Signed-off-by: Dongliang Cui <dongliang.cui@unisoc.com> --- block/blk-core.c | 1 + block/blk-mq-sched.c | 4 +++- block/blk-mq.c | 3 +++ block/blk-sysfs.c | 32 ++++++++++++++++++++++++++++++++ include/linux/blkdev.h | 2 ++ 5 files changed, 41 insertions(+), 1 deletion(-)