Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 2620:137:e000::1:20 as permitted sender) client-ip=2620:137:e000::1:20;
MIME-Version: 1.0
References: <20230530023227.16653-1-powen.kao@mediatek.com>
 <20230530023227.16653-2-powen.kao@mediatek.com> <1c182151-6e8c-5068-b38c-f8e842e6e13b@acm.org>
In-Reply-To: <1c182151-6e8c-5068-b38c-f8e842e6e13b@acm.org>
From:   Stanley Chu <chu.stanley@gmail.com>
Date:   Wed, 31 May 2023 09:54:25 +0800
Message-ID: <CAGaU9a9TsoaUtoMX8cNmAhfBnCddRtWYEG8ACdRNJfri+sdasg@mail.gmail.com>
Subject: Re: [PATCH v2 1/3] scsi: ufs: core: Introduce mcq ops to config cqid
To:     Bart Van Assche <bvanassche@acm.org>
Cc:     Po-Wen Kao <powen.kao@mediatek.com>, linux-scsi@vger.kernel.org,
        linux-kernel@vger.kernel.org, linux-arm-kernel@lists.infradead.org,
        linux-mediatek@lists.infradead.org,
        Alim Akhtar <alim.akhtar@samsung.com>,
        Avri Altman <avri.altman@wdc.com>,
        "James E.J. Bottomley" <jejb@linux.ibm.com>,
        "Martin K. Petersen" <martin.petersen@oracle.com>,
        Matthias Brugger <matthias.bgg@gmail.com>,
        AngeloGioacchino Del Regno 
        <angelogioacchino.delregno@collabora.com>,
        wsd_upstream@mediatek.com, peter.wang@mediatek.com,
        stanley.chu@mediatek.com, alice.chao@mediatek.com,
        naomi.chu@mediatek.com, chun-hung.wu@mediatek.com,
        cc.chou@mediatek.com, eddie.huang@mediatek.com
Content-Type: text/plain; charset="UTF-8"
Content-Transfer-Encoding: quoted-printable
Precedence: bulk

On Wed, May 31, 2023 at 7:54=E2=80=AFAM Bart Van Assche <bvanassche@acm.org=
> wrote:
>
> On 5/29/23 19:32, Po-Wen Kao wrote:
> > MCQ sq/cq mapping is not just one for one, could many for one.
> > This patch allow host driver to change the mapping, assign cqid
> > for each hw queue.
>
> What use case do you have in mind for associating multiple submission
> queues with a single completion queue?
>

According to the specification or the original concept of MCQ, the
bindings between SQ and CQ can not be fixed to a single style.
In addition, some benchmark data shows that the performance can be
improved by using fewer CQs to aggregate the interrupt handling of
completion requests.
Therefore, we would like to introduce a vop to allow the host to
configure it accordingly.

Stanley