Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]


Groups > linux.kernel > #1185940

Re: [PATCH 0/4] block: account io: kill atomic operations

From Jens Axboe <axboe@kernel.dk>
Newsgroups linux.kernel
Subject Re: [PATCH 0/4] block: account io: kill atomic operations
Date 2015-07-16 17:10 +0200
Message-ID <pMSXn-1Ro-5@gated-at.bofh.it> (permalink)
References <pMHSh-2E1-1@gated-at.bofh.it> <pMSE2-1eX-17@gated-at.bofh.it> <pMSXn-1Ro-7@gated-at.bofh.it>
Organization linux.* mail to news gateway

Show all headers | View raw


On 07/16/2015 09:01 AM, Ming Lei wrote:
> On Thu, Jul 16, 2015 at 10:48 PM, Jens Axboe <axboe@kernel.dk> wrote:
>> On 07/15/2015 09:16 PM, Ming Lei wrote:
>>>
>>> Hi,
>>>
>>> This patches kills two kinds of atomic operations in block
>>> accounting I/O.
>>>
>>> The 1st two patches convert atomic refcount of partition
>>> into percpu refcount.
>>>
>>> The 2nd two patches converts partition->in_flight[] into percpu
>>> variable.
>>>
>>> With this change, ~15% throughput improvement can be observed
>>> when running fio(randread) over null blk in a dual-socket
>>> environment.
>>
>>
>> I've played with this before, but always ran into the hurdle of making
>> part_in_flight() too expensive ended up hurting results in the end. Making
>
> Yes, it is a bit expensive, but it is only run at most one time per tick for
> one partition.

Yup, but that can still be 1000 per second. And up until last year, it 
was even worse: 7276d02e241dc. So it's not too surprising if there's 
more low hanging fruit :-)

If we can make the rounding more lazy, then we should go ahead and do that.

>> the inc/dec parts of accounting percpu is a no-brainer, unfortunately the
>> summing then becomes pretty expensive. I'll run this through some testing
>> and see what kind of results I get.
>
> The first two patches should be fine, and it still can get ~8% improvement
> in my test.

Agree, those can go right in.

-- 
Jens Axboe

--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/

Back to linux.kernel | Previous | NextPrevious in thread | Next in thread | Find similar | Unroll thread


Thread

[PATCH 0/4] block: account io: kill atomic operations Ming Lei <tom.leiming@gmail.com> - 2015-07-16 05:20 +0200
  [PATCH 3/4] block: partition: introduce 'cpu' para to part_inc|dec_in_flight Ming Lei <tom.leiming@gmail.com> - 2015-07-16 05:30 +0200
  [PATCH 1/4] block: partition: introduce hd_free_part() Ming Lei <tom.leiming@gmail.com> - 2015-07-16 05:30 +0200
  [PATCH 2/4] block: partition: convert percpu ref Ming Lei <tom.leiming@gmail.com> - 2015-07-16 05:30 +0200
    Re: [PATCH 2/4] block: partition: convert percpu ref Tejun Heo <tj@kernel.org> - 2015-07-16 16:40 +0200
  [PATCH 4/4] block: account io: convert part->in_fligh[] into percpu variable Ming Lei <tom.leiming@gmail.com> - 2015-07-16 05:30 +0200
    Re: [PATCH 4/4] block: account io: convert part->in_fligh[] into  percpu variable Tejun Heo <tj@kernel.org> - 2015-07-16 16:50 +0200
  Re: [PATCH 0/4] block: account io: kill atomic operations Jens Axboe <axboe@kernel.dk> - 2015-07-16 16:50 +0200
    Re: [PATCH 0/4] block: account io: kill atomic operations Tejun Heo <tj@kernel.org> - 2015-07-16 17:00 +0200
      Re: [PATCH 0/4] block: account io: kill atomic operations Jens Axboe <axboe@kernel.dk> - 2015-07-16 17:10 +0200
    Re: [PATCH 0/4] block: account io: kill atomic operations Jens Axboe <axboe@kernel.dk> - 2015-07-16 17:10 +0200
    Re: [PATCH 0/4] block: account io: kill atomic operations Ming Lei <tom.leiming@gmail.com> - 2015-07-16 17:10 +0200

csiph-web