Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 23.128.96.18 as permitted sender) client-ip=23.128.96.18;
Content-Type: text/plain;
        charset=us-ascii
Mime-Version: 1.0 (Mac OS X Mail 12.4 \(3445.104.11\))
Subject: Re: [PATCH BUGFIX 1/1] block, bfq: honor already-setup queue merges
From:   Paolo Valente <paolo.valente@linaro.org>
In-Reply-To: <20210802141352.74353-2-paolo.valente@linaro.org>
Date:   Thu, 26 Aug 2021 11:16:17 +0200
Cc:     linux-block@vger.kernel.org, linux-kernel@vger.kernel.org,
        davidezini2@gmail.com
Content-Transfer-Encoding: quoted-printable
Message-Id: <3BDD27BD-F49B-43B6-B41D-F7534B596A9A@linaro.org>
References: <20210802141352.74353-1-paolo.valente@linaro.org>
 <20210802141352.74353-2-paolo.valente@linaro.org>
To:     Jens Axboe <axboe@kernel.dk>
Precedence: bulk


> Il giorno 2 ago 2021, alle ore 16:13, Paolo Valente =
<paolo.valente@linaro.org> ha scritto:
>=20
> The function bfq_setup_merge prepares the merging between two
> bfq_queues, say bfqq and new_bfqq. To this goal, it assigns
> bfqq->new_bfqq =3D new_bfqq. Then, each time some I/O for bfqq =
arrives,
> the process that generated that I/O is disassociated from bfqq and
> associated with new_bfqq (merging is actually a redirection). In this
> respect, bfq_setup_merge increases new_bfqq->ref in advance, adding
> the number of processes that are expected to be associated with
> new_bfqq.
>=20
> Unfortunately, the stable-merging mechanism interferes with this
> setup. After bfqq->new_bfqq has been set by bfq_setup_merge, and
> before all the expected processes have been associated with
> bfqq->new_bfqq, bfqq may happen to be stably merged with a different
> queue than the current bfqq->new_bfqq. In this case, bfqq->new_bfqq
> gets changed. So, some of the processes that have been already
> accounted for in the ref counter of the previous new_bfqq will not be
> associated with that queue.  This creates an unbalance, because those
> references will never be decremented.
>=20
> This commit fixes this issue by reestablishing the previous, natural
> behaviour: once bfqq->new_bfqq has been set, it will not be changed
> until all expected redirections have occurred.
>=20

Hi Jens,
did you have time to look at this fix?

Thanks,
Paolo


> Signed-off-by: Davide Zini <davidezini2@gmail.com>
> Signed-off-by: Paolo Valente <paolo.valente@linaro.org>
> ---
> block/bfq-iosched.c | 16 +++++++++++++---
> 1 file changed, 13 insertions(+), 3 deletions(-)
>=20
> diff --git a/block/bfq-iosched.c b/block/bfq-iosched.c
> index 727955918563..08d9122dd4c0 100644
> --- a/block/bfq-iosched.c
> +++ b/block/bfq-iosched.c
> @@ -2659,6 +2659,15 @@ bfq_setup_merge(struct bfq_queue *bfqq, struct =
bfq_queue *new_bfqq)
> 	 * are likely to increase the throughput.
> 	 */
> 	bfqq->new_bfqq =3D new_bfqq;
> +	/*
> +	 * The above assignment schedules the following redirections:
> +	 * each time some I/O for bfqq arrives, the process that
> +	 * generated that I/O is disassociated from bfqq and
> +	 * associated with new_bfqq. Here we increases new_bfqq->ref
> +	 * in advance, adding the number of processes that are
> +	 * expected to be associated with new_bfqq as they happen to
> +	 * issue I/O.
> +	 */
> 	new_bfqq->ref +=3D process_refs;
> 	return new_bfqq;
> }
> @@ -2721,6 +2730,10 @@ bfq_setup_cooperator(struct bfq_data *bfqd, =
struct bfq_queue *bfqq,
> {
> 	struct bfq_queue *in_service_bfqq, *new_bfqq;
>=20
> +	/* if a merge has already been setup, then proceed with that =
first */
> +	if (bfqq->new_bfqq)
> +		return bfqq->new_bfqq;
> +
> 	/*
> 	 * Check delayed stable merge for rotational or non-queueing
> 	 * devs. For this branch to be executed, bfqq must not be
> @@ -2822,9 +2835,6 @@ bfq_setup_cooperator(struct bfq_data *bfqd, =
struct bfq_queue *bfqq,
> 	if (bfq_too_late_for_merging(bfqq))
> 		return NULL;
>=20
> -	if (bfqq->new_bfqq)
> -		return bfqq->new_bfqq;
> -
> 	if (!io_struct || unlikely(bfqq =3D=3D &bfqd->oom_bfqq))
> 		return NULL;
>=20
> --=20
> 2.20.1
>=20