On Fri, Sep 2, 2022 at 11:08 PM Yu Kuai [off-list ref] wrote:

Hi, Song

在 2022/09/01 2:00, Song Liu 写道:

quoted

On Mon, Aug 29, 2022 at 6:03 AM Yu Kuai [off-list ref] wrote:

quoted

From: Yu Kuai <redacted>

patch 1 is a small problem found by code review.
patch 2 avoid holding resync_lock in fast path.
patch 3 avoid holding lock in wake_up() in fast path.

Test environment:

Architecture: aarch64
Cpu: Huawei KUNPENG 920, there are four numa nodes

Raid10 initialize:
mdadm --create /dev/md0 --level 10 --bitmap none --raid-devices 4 /dev/nvme0n1 /dev/nvme1n1 /dev/nvme2n1 /dev/nvme3n1

Test cmd:
fio -name=0 -ioengine=libaio -direct=1 -group_reporting=1 -randseed=2022 -rwmixread=70 -refill_buffers -filename=/dev/md0 -numjobs=16 -runtime=60s -bs=4k -iodepth=256 -rw=randread

Test result:
before this patchset:   2.9 GiB/s
after this patchset:    6.6 Gib/s

Nice work! Applied to md-next.

Can you drop this version? There are something to improve. I can send a
new version.

Sure, I will drop it from md-next.

Song