Re: cleanup btrfs bio handling, part 2 v3

[Anand Jain <anand.jain@oracle.com>]

On 5/12/22 11:52, Anand Jain wrote:
> On 5/5/22 20:41, Christoph Hellwig wrote:
>> On Thu, May 05, 2022 at 02:56:49PM +0800, Qu Wenruo wrote:
>>> So do you mind to do some basic benchmark for read and write and see how
>>> the throughput and latency changed?
> 
> 
>> I'm not actually seeing any meaningful differences at all.  I think this
>> will help btrfs to better behave under memory pressure, though.
> This patch got me curious a couple of days back while I was tracing a
> dio read performance issue on nvme.
> 
> Yep. No meaningful difference. I am sharing the results as below. [1].
> 
> [1]
> Before:
>   4971MB/s 4474GB iocounts: nvme3n1 545220968 nvme0n1 547007640 
> single_d2/5.18.0-rc5+_misc-next_1
> 
> After:
>   4968MB/s 4471GB iocounts: nvme3n1 544207371 nvme1n1 547458037 
> single_d2/5.18.0-rc5+_dio_cleanup_hch_1
> 
>   readstat /btrfs fio --eta=auto --output=$CMDLOG \
> --name fiotest --directory=/btrfs --rw=randread \
> --bs=4k --size=4G --ioengine=libaio --iodepth=16 --direct=1 \
> --time_based=1 --runtime=900 --randrepeat=1 --gtod_reduce=1 \
> --group_reporting=1 --numjobs=64

  Oops wrong thread pls ignore.