Re: [PATCH v2 4/4] block: loop: support to submit I/O via kernel aio based

From: Ming Lei
Date: Wed Mar 18 2015 - 22:57:27 EST

On Thu, Mar 19, 2015 at 2:28 AM, Maxim Patlasov <mpatlasov@xxxxxxxxxxxxx> wrote:
> On 01/13/2015 07:44 AM, Ming Lei wrote:
>> Part of the patch is based on Dave's previous post.
>> This patch submits I/O to fs via kernel aio, and we
>> can obtain following benefits:
>> - double cache in both loop file system and backend file
>> gets avoided
>> - context switch decreased a lot, and finally CPU utilization
>> is decreased
>> - cached memory got decreased a lot
>> One main side effect is that throughput is decreased when
>> accessing raw loop block(not by filesystem) with kernel aio.
>> This patch has passed xfstests test(./check -g auto), and
>> both test and scratch devices are loop block, file system is ext4.
>> Follows two fio tests' result:
>> 1. fio test inside ext4 file system over loop block
>> 1) How to run
>> - linux kernel base: 3.19.0-rc3-next-20150108(loop-mq merged)
>> - loop over SSD image 1 in ext4
>> - linux psync, 16 jobs, size 200M, ext4 over loop block
>> - test result: IOPS from fio output
>> 2) Throughput result:
>> -------------------------------------------------------------
>> test cases |randread |read |randwrite |write |
>> -------------------------------------------------------------
>> base |16799 |59508 |31059 |58829
>> -------------------------------------------------------------
>> base+kernel aio |15480 |64453 |30187 |57222
>> -------------------------------------------------------------
> Ming, it's important to understand the overhead of aio_kernel_()
> implementation. So could you please add test results for raw SSD device to
> the table above next time (in v3 of your patches).

what aio_kernel_() does is to just call ->read_iter()/->write_iter(),
so it should not have introduced extra overload.

>From performance view, the effect is only from switching to
O_DIRECT. With O_DIRECT, double cache can be avoided,
meantime both page caches and CPU utilization can be decreased.

V3 need to rewrite against recent vfs's change(in -next already)
about splitting kiocb patches, and thanks to Christoph's patches,
aio_kernel_() can be implemented just inside loop and it may
become a bit simple.

> Jens, if you have some fast storage at hand, could you please measure IOPS
> for Ming's patches vs. raw block device -- to ensure that the patches do not
> impose too low limit on performance.
> Thanks,
> Maxim
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at
Please read the FAQ at