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: <20230705114505.63274-1-maxime.coquelin@redhat.com>
In-Reply-To: <20230705114505.63274-1-maxime.coquelin@redhat.com>
From:   Jason Wang <jasowang@redhat.com>
Date:   Wed, 26 Jul 2023 13:38:22 +0800
Message-ID: <CACGkMEtF9c0dL+bk0=JovcVs-ZVzEJJXdt9gx=_Lh+KtwFu9ig@mail.gmail.com>
Subject: Re: [PATCH] vduse: Use proper spinlock for IRQ injection
To:     Maxime Coquelin <maxime.coquelin@redhat.com>
Cc:     xieyongji@bytedance.com, mst@redhat.com, david.marchand@redhat.com,
        lulu@redhat.com, linux-kernel@vger.kernel.org,
        virtualization@lists.linux-foundation.org, netdev@vger.kernel.org,
        xuanzhuo@linux.alibaba.com, eperezma@redhat.com
Content-Type: text/plain; charset="UTF-8"
Content-Transfer-Encoding: quoted-printable
Precedence: bulk

On Wed, Jul 5, 2023 at 7:45=E2=80=AFPM Maxime Coquelin
<maxime.coquelin@redhat.com> wrote:
>
> The IRQ injection work used spin_lock_irq() to protect the
> scheduling of the softirq, but spin_lock_bh() should be
> used.
>
> With spin_lock_irq(), we noticed delay of more than 6
> seconds between the time a NAPI polling work is scheduled
> and the time it is executed.
>
> Fixes: c8a6153b6c59 ("vduse: Introduce VDUSE - vDPA Device in Userspace")
> Cc: xieyongji@bytedance.com
>
> Suggested-by: Jason Wang <jasowang@redhat.com>
> Signed-off-by: Maxime Coquelin <maxime.coquelin@redhat.com>

Acked-by: Jason Wang <jasowang@redhat.com>

Thanks


> ---
>  drivers/vdpa/vdpa_user/vduse_dev.c | 8 ++++----
>  1 file changed, 4 insertions(+), 4 deletions(-)
>
> diff --git a/drivers/vdpa/vdpa_user/vduse_dev.c b/drivers/vdpa/vdpa_user/=
vduse_dev.c
> index dc38ed21319d..df7869537ef1 100644
> --- a/drivers/vdpa/vdpa_user/vduse_dev.c
> +++ b/drivers/vdpa/vdpa_user/vduse_dev.c
> @@ -935,10 +935,10 @@ static void vduse_dev_irq_inject(struct work_struct=
 *work)
>  {
>         struct vduse_dev *dev =3D container_of(work, struct vduse_dev, in=
ject);
>
> -       spin_lock_irq(&dev->irq_lock);
> +       spin_lock_bh(&dev->irq_lock);
>         if (dev->config_cb.callback)
>                 dev->config_cb.callback(dev->config_cb.private);
> -       spin_unlock_irq(&dev->irq_lock);
> +       spin_unlock_bh(&dev->irq_lock);
>  }
>
>  static void vduse_vq_irq_inject(struct work_struct *work)
> @@ -946,10 +946,10 @@ static void vduse_vq_irq_inject(struct work_struct =
*work)
>         struct vduse_virtqueue *vq =3D container_of(work,
>                                         struct vduse_virtqueue, inject);
>
> -       spin_lock_irq(&vq->irq_lock);
> +       spin_lock_bh(&vq->irq_lock);
>         if (vq->ready && vq->cb.callback)
>                 vq->cb.callback(vq->cb.private);
> -       spin_unlock_irq(&vq->irq_lock);
> +       spin_unlock_bh(&vq->irq_lock);
>  }
>
>  static bool vduse_vq_signal_irqfd(struct vduse_virtqueue *vq)
> --
> 2.41.0
>