[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <513db882-ebe1-7287-99dd-3783c2c9b90f@huaweicloud.com>
Date: Sat, 6 Apr 2024 15:57:14 +0800
From: Yu Kuai <yukuai1@...weicloud.com>
To: Yu Kuai <yukuai1@...weicloud.com>, ming.lei@...hat.com, hch@....de,
bvanassche@....org, axboe@...nel.dk, mpatocka@...hat.com, snitzer@...hat.com
Cc: linux-block@...r.kernel.org, linux-kernel@...r.kernel.org,
yi.zhang@...wei.com, yangerkun@...wei.com, "yukuai (C)" <yukuai3@...wei.com>
Subject: Re: [PATCH 0/2] block: support to account io_ticks precisely
Hi, Jens!
Hi, Ming!
Hi, Christoph!
Hi, Bart!
Friendly ping ...
The 'util' reported by iostat is very important for users, they don't
have much choise to get disk status, while 'util' has been inaccurate
for a long time unnecessarily. I really think patch 1 is meaningful.
Patch 2 also tries to fix a problem by our customer that util can
sometimes be huge. The root cause is that 'inflight' is account from
blk_mq_start_request() while 'io_ticks' is account from
blk_account_io_start(), there is a gap. I let 'inflight' to be account
from blk_account_io_start() as well, please let me know if this is not
good.
Thanks!
Kuai
在 2024/03/23 11:59, Yu Kuai 写道:
> From: Yu Kuai <yukuai3@...wei.com>
>
> Yu Kuai (2):
> block: support to account io_ticks precisely
> block: remove blk_mq_in_flight() and blk_mq_in_flight_rw()
>
> block/blk-core.c | 9 +++++----
> block/blk-merge.c | 2 ++
> block/blk-mq.c | 36 ++++--------------------------------
> block/blk-mq.h | 5 -----
> block/blk.h | 1 +
> block/genhd.c | 20 ++++----------------
> 6 files changed, 16 insertions(+), 57 deletions(-)
>
Powered by blists - more mailing lists