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: <1664528184-162714-1-git-send-email-john.garry@huawei.com>
 <1664528184-162714-5-git-send-email-john.garry@huawei.com>
 <CAMGffEkgR6VmfMV1eLVeTn+b8b0GC_B9=vj3HxF6cn5Er6+CTg@mail.gmail.com>
 <5e570c35-ed98-a29b-736f-677aa896f654@huawei.com> <32b0f929-4534-450b-b19c-d3404db63202@huawei.com>
In-Reply-To: <32b0f929-4534-450b-b19c-d3404db63202@huawei.com>
From:   Jinpu Wang <jinpu.wang@ionos.com>
Date:   Tue, 4 Oct 2022 06:51:35 +0200
Message-ID: <CAMGffEkPnrjjzZMca5YQ8NT-xnBFPfFoBxs63jY431FtpFvF3A@mail.gmail.com>
Subject: Re: [PATCH v2 4/6] scsi: pm8001: Use sas_task_find_rq() for tagging
To:     John Garry <john.garry@huawei.com>
Cc:     damien.lemoal@wdc.com, jejb@linux.ibm.com,
        martin.petersen@oracle.com, hare@suse.de,
        linux-scsi@vger.kernel.org, linux-kernel@vger.kernel.org,
        linuxarm@huawei.com, ipylypiv@google.com, changyuanl@google.com,
        hch@lst.de, yanaijie@huawei.com
Content-Type: text/plain; charset="UTF-8"
Precedence: bulk

On Fri, Sep 30, 2022 at 1:05 PM John Garry <john.garry@huawei.com> wrote:
>
> On 30/09/2022 11:20, John Garry wrote:
> > BTW, I do notice that we still have this global lock in delivery path
> > which should be removed at some stage:
> > > int mvs_queue_command(struct sas_task *task, gfp_t gfp_flags)
> > {
> >      ...
> >
> >      spin_lock_irqsave(&mvi->lock, flags);
> >      rc = mvs_task_prep(task, mvi, is_tmf, tmf, &pass);
> >      ...
> >      spin_unlock_irqrestore(&mvi->lock, flags);
> > }
> >
>
> oops... that's mvsas. But pm8001 does still use a big lock (which we
> should get rid off):
Yes, would be great to get rid of the per ha lock.
>
> int pm8001_queue_command(struct sas_task *task, gfp_t gfp_flags)
> {
>         ...
>         pm8001_dbg(pm8001_ha, IO, "pm8001_task_exec device\n");
>
>         spin_lock_irqsave(&pm8001_ha->lock, flags);
>
>
> Thanks,
> John
>
> > That really will affect performance...
>