Re: [PATCH -next v5 0/3] support concurrent sync io for bfq on a specail occasion

["yukuai (C)" <yukuai3@huawei.com>]

在 2022/05/05 9:00, yukuai (C) 写道:
> Hi, Paolo
> 
> Can you take a look at this patchset? It has been quite a long time
> since we spotted this problem...
> 

friendly ping ...
> Thanks,
> Kuai
> 
> 在 2022/04/28 20:08, Yu Kuai 写道:
>> Changes in v5:
>>   - rename bfq_add_busy_queues() to bfq_inc_busy_queues() in patch 1
>>   - fix wrong definition in patch 1
>>   - fix spelling mistake in patch 2: leaset -> least
>>   - update comments in patch 3
>>   - add reviewed-by tag in patch 2,3
>>
>> Changes in v4:
>>   - split bfq_update_busy_queues() to bfq_add/dec_busy_queues(),
>>     suggested by Jan Kara.
>>   - remove unused 'in_groups_with_pending_reqs',
>>
>> Changes in v3:
>>   - remove the cleanup patch that is irrelevant now(I'll post it
>>     separately).
>>   - instead of hacking wr queues and using weights tree 
>> insertion/removal,
>>     using bfq_add/del_bfqq_busy() to count the number of groups
>>     (suggested by Jan Kara).
>>
>> Changes in v2:
>>   - Use a different approch to count root group, which is much simple.
>>
>> Currently, bfq can't handle sync io concurrently as long as they
>> are not issued from root group. This is because
>> 'bfqd->num_groups_with_pending_reqs > 0' is always true in
>> bfq_asymmetric_scenario().
>>
>> The way that bfqg is counted into 'num_groups_with_pending_reqs':
>>
>> Before this patchset:
>>   1) root group will never be counted.
>>   2) Count if bfqg or it's child bfqgs have pending requests.
>>   3) Don't count if bfqg and it's child bfqgs complete all the requests.
>>
>> After this patchset:
>>   1) root group is counted.
>>   2) Count if bfqg have at least one bfqq that is marked busy.
>>   3) Don't count if bfqg doesn't have any busy bfqqs.
>>
>> The main reason to use busy state of bfqq instead of 'pending requests'
>> is that bfqq can stay busy after dispatching the last request if idling
>> is needed for service guarantees.
>>
>> With the above changes, concurrent sync io can be supported if only
>> one group is activated.
>>
>> fio test script(startdelay is used to avoid queue merging):
>> [global]
>> filename=/dev/nvme0n1
>> allow_mounted_write=0
>> ioengine=psync
>> direct=1
>> ioscheduler=bfq
>> offset_increment=10g
>> group_reporting
>> rw=randwrite
>> bs=4k
>>
>> [test1]
>> numjobs=1
>>
>> [test2]
>> startdelay=1
>> numjobs=1
>>
>> [test3]
>> startdelay=2
>> numjobs=1
>>
>> [test4]
>> startdelay=3
>> numjobs=1
>>
>> [test5]
>> startdelay=4
>> numjobs=1
>>
>> [test6]
>> startdelay=5
>> numjobs=1
>>
>> [test7]
>> startdelay=6
>> numjobs=1
>>
>> [test8]
>> startdelay=7
>> numjobs=1
>>
>> test result:
>> running fio on root cgroup
>> v5.18-rc1:       550 Mib/s
>> v5.18-rc1-patched: 550 Mib/s
>>
>> running fio on non-root cgroup
>> v5.18-rc1:       349 Mib/s
>> v5.18-rc1-patched: 550 Mib/s
>>
>> Note that I also test null_blk with "irqmode=2
>> completion_nsec=100000000(100ms) hw_queue_depth=1", and tests show
>> that service guarantees are still preserved.
>>
>> Previous versions:
>> RFC: 
>> https://lore.kernel.org/all/20211127101132.486806-1-yukuai3@huawei.com/
>> v1: 
>> https://lore.kernel.org/all/20220305091205.4188398-1-yukuai3@huawei.com/
>> v2: 
>> https://lore.kernel.org/all/20220416093753.3054696-1-yukuai3@huawei.com/
>> v3: 
>> https://lore.kernel.org/all/20220427124722.48465-1-yukuai3@huawei.com/
>> v4: 
>> https://lore.kernel.org/all/20220428111907.3635820-1-yukuai3@huawei.com/
>>
>> Yu Kuai (3):
>>    block, bfq: record how many queues are busy in bfq_group
>>    block, bfq: refactor the counting of 'num_groups_with_pending_reqs'
>>    block, bfq: do not idle if only one group is activated
>>
>>   block/bfq-cgroup.c  |  1 +
>>   block/bfq-iosched.c | 48 +++-----------------------------------
>>   block/bfq-iosched.h | 57 +++++++--------------------------------------
>>   block/bfq-wf2q.c    | 35 +++++++++++++++++-----------
>>   4 files changed, 35 insertions(+), 106 deletions(-)
>>