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] Revert "bfq: Fix computation of shallow depth"
From:   Paolo Valente <paolo.valente@linaro.org>
In-Reply-To: <749df7e1-98b5-a575-506d-a7c7c4d96a6f@wangsu.com>
Date:   Mon, 1 Feb 2021 20:03:02 +0100
Cc:     Jens Axboe <axboe@kernel.dk>, Jan Kara <jack@suse.cz>,
        Linux Kernel Mailing List <linux-kernel@vger.kernel.org>,
        linux-block@vger.kernel.org
Content-Transfer-Encoding: quoted-printable
Message-Id: <5CCEC0EB-60DA-43C3-A251-BDB96B036B09@linaro.org>
References: <20210129111808.45796-1-linf@wangsu.com>
 <749df7e1-98b5-a575-506d-a7c7c4d96a6f@wangsu.com>
To:     Lin Feng <linf@wangsu.com>
Precedence: bulk


> Il giorno 1 feb 2021, alle ore 08:32, Lin Feng <linf@wangsu.com> ha =
scritto:
>=20
> Hi, it seems that this patch was blocked by linux mailist servers, so =
ping again.
>=20
> Based on =
https://patchwork.kernel.org/project/linux-block/patch/20201210094433.2549=
1-1-jack@suse.cz/,
> it looks like we have made a consensus about bfqd->word_depths[2][2]'s =
changing, so now the
> computation codes for bfq's word_depths array are not necessary and =
one variable is enough.
>=20
> But IMHO async depth limitation for slow drivers is essential, which =
is what we always did in cfq age.
>=20

It is essential.

Thanks,
Paolo

> On 1/29/21 19:18, Lin Feng wrote:
>> This reverts commit 6d4d273588378c65915acaf7b2ee74e9dd9c130a.
>> bfq.limit_depth passes word_depths[] as shallow_depth down to sbitmap =
core
>> sbitmap_get_shallow, which uses just the number to limit the scan =
depth of
>> each bitmap word, formula:
>> scan_percentage_for_each_word =3D shallow_depth / (1 << =
sbimap->shift) * 100%
>> That means the comments's percentiles 50%, 75%, 18%, 37% of bfq are =
correct.
>> But after commit patch 'bfq: Fix computation of shallow depth', we =
use
>> sbitmap.depth instead, as a example in following case:
>> sbitmap.depth =3D 256, map_nr =3D 4, shift =3D 6; sbitmap_word.depth =
=3D 64.
>> The resulsts of computed bfqd->word_depths[] are {128, 192, 48, 96}, =
and
>> three of the numbers exceed core dirver's 'sbitmap_word.depth=3D64' =
limit
>> nothing. Do we really don't want limit depth for such workloads, or =
we
>> just want to bump up the percentiles to 100%?
>> Please correct me if I miss something, thanks.
>> Signed-off-by: Lin Feng <linf@wangsu.com>
>> ---
>>  block/bfq-iosched.c | 8 ++++----
>>  1 file changed, 4 insertions(+), 4 deletions(-)
>> diff --git a/block/bfq-iosched.c b/block/bfq-iosched.c
>> index 9e4eb0fc1c16..9e81d1052091 100644
>> --- a/block/bfq-iosched.c
>> +++ b/block/bfq-iosched.c
>> @@ -6332,13 +6332,13 @@ static unsigned int bfq_update_depths(struct =
bfq_data *bfqd,
>>  	 * limit 'something'.
>>  	 */
>>  	/* no more than 50% of tags for async I/O */
>> -	bfqd->word_depths[0][0] =3D max(bt->sb.depth >> 1, 1U);
>> +	bfqd->word_depths[0][0] =3D max((1U << bt->sb.shift) >> 1, 1U);
>>  	/*
>>  	 * no more than 75% of tags for sync writes (25% extra tags
>>  	 * w.r.t. async I/O, to prevent async I/O from starving sync
>>  	 * writes)
>>  	 */
>> -	bfqd->word_depths[0][1] =3D max((bt->sb.depth * 3) >> 2, 1U);
>> +	bfqd->word_depths[0][1] =3D max(((1U << bt->sb.shift) * 3) >> 2, =
1U);
>>    	/*
>>  	 * In-word depths in case some bfq_queue is being weight-
>> @@ -6348,9 +6348,9 @@ static unsigned int bfq_update_depths(struct =
bfq_data *bfqd,
>>  	 * shortage.
>>  	 */
>>  	/* no more than ~18% of tags for async I/O */
>> -	bfqd->word_depths[1][0] =3D max((bt->sb.depth * 3) >> 4, 1U);
>> +	bfqd->word_depths[1][0] =3D max(((1U << bt->sb.shift) * 3) >> 4, =
1U);
>>  	/* no more than ~37% of tags for sync writes (~20% extra tags) =
*/
>> -	bfqd->word_depths[1][1] =3D max((bt->sb.depth * 6) >> 4, 1U);
>> +	bfqd->word_depths[1][1] =3D max(((1U << bt->sb.shift) * 6) >> 4, =
1U);
>>    	for (i =3D 0; i < 2; i++)
>>  		for (j =3D 0; j < 2; j++)
>=20