From: Jan Kara <jack@suse.cz>
To: Kemeng Shi <shikemeng@huaweicloud.com>
Cc: paolo.valente@linaro.org, axboe@kernel.dk,
linux-block@vger.kernel.org, linux-kernel@vger.kernel.org,
jack@suse.cz, hch@lst.de, damien.lemoal@wdc.com
Subject: Re: [PATCH RESEND v2 03/10] block, bfq: initialize bfqq->decrease_time_jif correctly
Date: Mon, 2 Jan 2023 16:30:56 +0100 [thread overview]
Message-ID: <20230102153056.gdofu7rp5s4arz2n@quack3> (raw)
In-Reply-To: <20221222191641.1643117-4-shikemeng@huaweicloud.com>
On Fri 23-12-22 03:16:34, Kemeng Shi wrote:
> Inject limit is updated or reset when time_is_before_eq_jiffies(
> decrease_time_jif + several msecs) or think-time state changes.
> decrease_time_jif is initialized to 0 and will be set to current jiffies
> when inject limit is updated or reset. If the jiffies is slightly greater
> than LONG_MAX, time_is_after_eq_jiffies(0) will keep for a long time, so as
> time_is_after_eq_jiffies(decrease_time_jif + several msecs). If the
> think-time state never chages, then the injection will not work as expected
> for long time.
>
> To be more specific:
> Function bfq_update_inject_limit maybe triggered when jiffies pasts
> decrease_time_jif + msecs_to_jiffies(10) in bfq_add_request by setting
> bfqd->wait_dispatch to true.
> Function bfq_reset_inject_limit are called in two conditions:
> 1. jiffies pasts bfqq->decrease_time_jif + msecs_to_jiffies(1000) in
> function bfq_add_request.
> 2. jiffies pasts bfqq->decrease_time_jif + msecs_to_jiffies(100) or
> bfq think-time state change from short to long.
>
> Fix this by initializing bfqq->decrease_time_jif to current jiffies
> to trigger service injection soon when service injection conditions
> are met.
>
> Signed-off-by: Kemeng Shi <shikemeng@huaweicloud.com>
I agree that initializing jiffies timer to 0 is just a bad practice and
current time is much saner choice. Feel free to add:
Reviewed-by: Jan Kara <jack@suse.cz>
Honza
> ---
> block/bfq-iosched.c | 2 ++
> 1 file changed, 2 insertions(+)
>
> diff --git a/block/bfq-iosched.c b/block/bfq-iosched.c
> index 5a6d9e8c329d..9441fc98961a 100644
> --- a/block/bfq-iosched.c
> +++ b/block/bfq-iosched.c
> @@ -5491,6 +5491,8 @@ static void bfq_init_bfqq(struct bfq_data *bfqd, struct bfq_queue *bfqq,
>
> /* first request is almost certainly seeky */
> bfqq->seek_history = 1;
> +
> + bfqq->decrease_time_jif = jiffies;
> }
>
> static struct bfq_queue **bfq_async_queue_prio(struct bfq_data *bfqd,
> --
> 2.30.0
>
--
Jan Kara <jack@suse.com>
SUSE Labs, CR
next prev parent reply other threads:[~2023-01-02 15:31 UTC|newest]
Thread overview: 21+ messages / expand[flat|nested] mbox.gz Atom feed top
2022-12-22 19:16 [PATCH RESEND v2 00/10] A few bugfix and cleancode patch for bfq Kemeng Shi
2022-12-22 19:16 ` [PATCH RESEND v2 01/10] block, bfq: correctly raise inject limit in bfq_choose_bfqq_for_injection Kemeng Shi
2023-01-02 15:13 ` Jan Kara
2022-12-22 19:16 ` [PATCH RESEND v2 02/10] block, bfq: remove unsed parameter reason in bfq_bfqq_is_slow Kemeng Shi
2023-01-02 15:14 ` Jan Kara
2022-12-22 19:16 ` [PATCH RESEND v2 03/10] block, bfq: initialize bfqq->decrease_time_jif correctly Kemeng Shi
2023-01-02 15:30 ` Jan Kara [this message]
2022-12-22 19:16 ` [PATCH RESEND v2 04/10] block, bfq: use helper macro RQ_BFQQ to get bfqq of request Kemeng Shi
2023-01-02 15:31 ` Jan Kara
2022-12-22 19:16 ` [PATCH RESEND v2 05/10] block, bfq: remove unnecessary dereference to get async_bfqq Kemeng Shi
2023-01-02 15:34 ` Jan Kara
2022-12-22 19:16 ` [PATCH RESEND v2 06/10] block, bfq: remove redundant bfqd->rq_in_driver > 0 check in bfq_add_request Kemeng Shi
2023-01-02 15:54 ` Jan Kara
2022-12-22 19:16 ` [PATCH RESEND v2 07/10] block, bfq: remove redundant check in bfq_put_cooperator Kemeng Shi
2023-01-02 15:57 ` Jan Kara
2022-12-22 19:16 ` [PATCH RESEND v2 08/10] block, bfq: remove unnecessary goto tag in bfq_dispatch_rq_from_bfqq Kemeng Shi
2023-01-02 15:59 ` Jan Kara
2022-12-22 19:16 ` [PATCH RESEND v2 09/10] block, bfq: remove unused bfq_wr_max_time in struct bfq_data Kemeng Shi
2023-01-02 16:00 ` Jan Kara
2022-12-22 19:16 ` [PATCH RESEND v2 10/10] block, bfq: remove check of bfq_wr_max_softrt_rate which is always greater than 0 Kemeng Shi
2023-01-02 16:05 ` Jan Kara
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=20230102153056.gdofu7rp5s4arz2n@quack3 \
--to=jack@suse.cz \
--cc=axboe@kernel.dk \
--cc=damien.lemoal@wdc.com \
--cc=hch@lst.de \
--cc=linux-block@vger.kernel.org \
--cc=linux-kernel@vger.kernel.org \
--cc=paolo.valente@linaro.org \
--cc=shikemeng@huaweicloud.com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox