Linux block layer
 help / color / mirror / Atom feed
From: Jan Kara <jack@suse.cz>
To: Jens Axboe <axboe@kernel.dk>
Cc: <linux-block@vger.kernel.org>,
	Damien Le Moal <damien.lemoal@opensource.wdc.com>,
	Niklas Cassel <Niklas.Cassel@wdc.com>,
	Bart Van Assche <bvanassche@acm.org>, Jan Kara <jack@suse.cz>
Subject: [PATCH 1/8] block: Fix handling of tasks without ioprio in ioprio_get(2)
Date: Wed, 15 Jun 2022 18:16:04 +0200	[thread overview]
Message-ID: <20220615161616.5055-1-jack@suse.cz> (raw)
In-Reply-To: <20220615160437.5478-1-jack@suse.cz>

ioprio_get(2) can be asked to return the best IO priority from several
tasks (IOPRIO_WHO_PGRP, IOPRIO_WHO_USER). Currently the call treats
tasks without set IO priority as having priority
IOPRIO_CLASS_BE/IOPRIO_BE_NORM however this does not really reflect the
IO priority the task will get (which depends on task's nice value) and
with the following fix it will not even match returned IO priority for a
single task.

Fix IO priority comparison to treat unset IO priority as the lowest
possible one. This way we will return IOPRIO_CLASS_NONE priority only if
none of the considered tasks has explicitely set IO priority, otherwise
we return the highest set IO priority. This changes userspace visible
behavior but this way the caller really gets back the highest set IO
priority without contaminating the result with
IOPRIO_CLASS_BE/IOPRIO_BE_NORM from tasks which have IO priority unset.

Signed-off-by: Jan Kara <jack@suse.cz>
---
 block/ioprio.c | 5 ++---
 1 file changed, 2 insertions(+), 3 deletions(-)

diff --git a/block/ioprio.c b/block/ioprio.c
index 2fe068fcaad5..62890391fc80 100644
--- a/block/ioprio.c
+++ b/block/ioprio.c
@@ -157,10 +157,9 @@ static int get_task_ioprio(struct task_struct *p)
 int ioprio_best(unsigned short aprio, unsigned short bprio)
 {
 	if (!ioprio_valid(aprio))
-		aprio = IOPRIO_DEFAULT;
+		return bprio;
 	if (!ioprio_valid(bprio))
-		bprio = IOPRIO_DEFAULT;
-
+		return aprio;
 	return min(aprio, bprio);
 }
 
-- 
2.35.3


  reply	other threads:[~2022-06-15 16:19 UTC|newest]

Thread overview: 20+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2022-06-15 16:16 [PATCH 0/8 v2] block: Fix IO priority mess Jan Kara
2022-06-15 16:16 ` Jan Kara [this message]
2022-06-15 16:16 ` [PATCH 2/8] block: Make ioprio_best() static Jan Kara
2022-06-15 16:16 ` [PATCH 3/8] block: fix default IO priority handling again Jan Kara
2022-06-15 16:16 ` [PATCH 4/8] block: Return effective IO priority from get_current_ioprio() Jan Kara
2022-06-15 16:16 ` [PATCH 5/8] blk-ioprio: Remove unneeded field Jan Kara
2022-06-15 16:16 ` [PATCH 6/8] blk-ioprio: Convert from rqos policy to direct call Jan Kara
2022-06-16  3:14   ` Damien Le Moal
2022-06-15 16:16 ` [PATCH 7/8] block: Initialize bio priority earlier Jan Kara
2022-06-15 16:16 ` [PATCH 8/8] block: Always initialize bio IO priority on submit Jan Kara
2022-06-16  3:15   ` Damien Le Moal
2022-06-16 11:23     ` Jan Kara
2022-06-16 12:24       ` Jan Kara
2022-06-17  0:04         ` Damien Le Moal
2022-06-17 11:49           ` Jan Kara
2022-06-17 12:03             ` Damien Le Moal
2022-06-17 14:54               ` Jan Kara
2022-06-17  0:05       ` Damien Le Moal
2022-06-17 11:52         ` Jan Kara
2022-06-17 12:00           ` Damien Le Moal

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=20220615161616.5055-1-jack@suse.cz \
    --to=jack@suse.cz \
    --cc=Niklas.Cassel@wdc.com \
    --cc=axboe@kernel.dk \
    --cc=bvanassche@acm.org \
    --cc=damien.lemoal@opensource.wdc.com \
    --cc=linux-block@vger.kernel.org \
    /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