Re: [linus:master] [mpage] 7d28631786: fio.write_iops 27.8% improvement

To: kernel test robot <yujie.liu@xxxxxxxxx>
Subject: Re: [linus:master] [mpage] 7d28631786: fio.write_iops 27.8% improvement
From: Christoph Hellwig <hch@xxxxxx>
Date: Wed, 15 Mar 2023 08:58:16 +0100
Cc: Christoph Hellwig <hch@xxxxxx>, oe-lkp@xxxxxxxxxxxxxxx, lkp@xxxxxxxxx, linux-kernel@xxxxxxxxxxxxxxx, Andrew Morton <akpm@xxxxxxxxxxxxxxxxxxxx>, Dan Williams <dan.j.williams@xxxxxxxxx>, Keith Busch <kbusch@xxxxxxxxxx>, Dave Jiang <dave.jiang@xxxxxxxxx>, Ira Weiny <ira.weiny@xxxxxxxxx>, Jens Axboe <axboe@xxxxxxxxx>, Minchan Kim <minchan@xxxxxxxxxx>, Sergey Senozhatsky <senozhatsky@xxxxxxxxxxxx>, Vishal Verma <vishal.l.verma@xxxxxxxxx>, linux-fsdevel@xxxxxxxxxxxxxxx, ying.huang@xxxxxxxxx, feng.tang@xxxxxxxxx, zhengjun.xing@xxxxxxxxxxxxxxx, fengwei.yin@xxxxxxxxx
In-reply-to: <202303101630.ef282023-yujie.liu@intel.com>
User-agent: Mutt/1.5.17 (2007-11-01)

I think this simply accounts for the I/Os now that were skipped
when using the bdev_read/write path before.

Follow-Ups:
- Re: [linus:master] [mpage] 7d28631786: fio.write_iops 27.8% improvement
  - From: Sergey Senozhatsky

References:
- [linus:master] [mpage] 7d28631786: fio.write_iops 27.8% improvement
  - From: kernel test robot