Message-ID: <CACycT3ucaY-QYq14BHf5oqLjQ0Nc9ZTQ0DA0ATu8=-KkaU5b3Q@mail.gmail.com>
Date:   Mon, 31 May 2021 14:24:29 +0800
From:   Yongji Xie <xieyongji@...edance.com>
To:     Jason Wang <jasowang@...hat.com>
Cc:     "Michael S. Tsirkin" <mst@...hat.com>,
        Stefan Hajnoczi <stefanha@...hat.com>,
        Stefano Garzarella <sgarzare@...hat.com>,
        Parav Pandit <parav@...dia.com>,
        Christoph Hellwig <hch@...radead.org>,
        Christian Brauner <christian.brauner@...onical.com>,
        Randy Dunlap <rdunlap@...radead.org>,
        Matthew Wilcox <willy@...radead.org>,
        Al Viro <viro@...iv.linux.org.uk>,
        Jens Axboe <axboe@...nel.dk>, bcrl@...ck.org,
        Jonathan Corbet <corbet@....net>,
        Mika Penttilä <mika.penttila@...tfour.com>,
        Dan Carpenter <dan.carpenter@...cle.com>, joro@...tes.org,
        virtualization <virtualization@...ts.linux-foundation.org>,
        netdev@...r.kernel.org, kvm <kvm@...r.kernel.org>,
        linux-fsdevel@...r.kernel.org, iommu@...ts.linux-foundation.org,
        linux-kernel <linux-kernel@...r.kernel.org>
Subject: Re: Re: [PATCH v7 11/12] vduse: Introduce VDUSE - vDPA Device in Userspace

On Mon, May 31, 2021 at 12:39 PM Jason Wang <jasowang@...hat.com> wrote:
>
>
> 在 2021/5/31 下午12:27, Yongji Xie 写道:
> > On Fri, May 28, 2021 at 10:31 AM Jason Wang <jasowang@...hat.com> wrote:
> >>
> >> 在 2021/5/27 下午9:17, Yongji Xie 写道:
> >>> On Thu, May 27, 2021 at 4:41 PM Jason Wang <jasowang@...hat.com> wrote:
> >>>> 在 2021/5/27 下午3:34, Yongji Xie 写道:
> >>>>> On Thu, May 27, 2021 at 1:40 PM Jason Wang <jasowang@...hat.com> wrote:
> >>>>>> 在 2021/5/27 下午1:08, Yongji Xie 写道:
> >>>>>>> On Thu, May 27, 2021 at 1:00 PM Jason Wang <jasowang@...hat.com> wrote:
> >>>>>>>> 在 2021/5/27 下午12:57, Yongji Xie 写道:
> >>>>>>>>> On Thu, May 27, 2021 at 12:13 PM Jason Wang <jasowang@...hat.com> wrote:
> >>>>>>>>>> 在 2021/5/17 下午5:55, Xie Yongji 写道:
> >>>>>>>>>>> +
> >>>>>>>>>>> +static int vduse_dev_msg_sync(struct vduse_dev *dev,
> >>>>>>>>>>> +                           struct vduse_dev_msg *msg)
> >>>>>>>>>>> +{
> >>>>>>>>>>> +     init_waitqueue_head(&msg->waitq);
> >>>>>>>>>>> +     spin_lock(&dev->msg_lock);
> >>>>>>>>>>> +     vduse_enqueue_msg(&dev->send_list, msg);
> >>>>>>>>>>> +     wake_up(&dev->waitq);
> >>>>>>>>>>> +     spin_unlock(&dev->msg_lock);
> >>>>>>>>>>> +     wait_event_killable(msg->waitq, msg->completed);
> >>>>>>>>>> What happens if the userspace(malicous) doesn't give a response forever?
> >>>>>>>>>>
> >>>>>>>>>> It looks like a DOS. If yes, we need to consider a way to fix that.
> >>>>>>>>>>
> >>>>>>>>> How about using wait_event_killable_timeout() instead?
> >>>>>>>> Probably, and then we need choose a suitable timeout and more important,
> >>>>>>>> need to report the failure to virtio.
> >>>>>>>>
> >>>>>>> Makes sense to me. But it looks like some
> >>>>>>> vdpa_config_ops/virtio_config_ops such as set_status() didn't have a
> >>>>>>> return value.  Now I add a WARN_ON() for the failure. Do you mean we
> >>>>>>> need to add some change for virtio core to handle the failure?
> >>>>>> Maybe, but I'm not sure how hard we can do that.
> >>>>>>
> >>>>> We need to change all virtio device drivers in this way.
> >>>> Probably.
> >>>>
> >>>>
> >>>>>> We had NEEDS_RESET but it looks we don't implement it.
> >>>>>>
> >>>>> Could it handle the failure of get_feature() and get/set_config()?
> >>>> Looks not:
> >>>>
> >>>> "
> >>>>
> >>>> The device SHOULD set DEVICE_NEEDS_RESET when it enters an error state
> >>>> that a reset is needed. If DRIVER_OK is set, after it sets
> >>>> DEVICE_NEEDS_RESET, the device MUST send a device configuration change
> >>>> notification to the driver.
> >>>>
> >>>> "
> >>>>
> >>>> This looks implies that NEEDS_RESET may only work after device is
> >>>> probed. But in the current design, even the reset() is not reliable.
> >>>>
> >>>>
> >>>>>> Or a rough idea is that maybe need some relaxing to be coupled loosely
> >>>>>> with userspace. E.g the device (control path) is implemented in the
> >>>>>> kernel but the datapath is implemented in the userspace like TUN/TAP.
> >>>>>>
> >>>>> I think it can work for most cases. One problem is that the set_config
> >>>>> might change the behavior of the data path at runtime, e.g.
> >>>>> virtnet_set_mac_address() in the virtio-net driver and
> >>>>> cache_type_store() in the virtio-blk driver. Not sure if this path is
> >>>>> able to return before the datapath is aware of this change.
> >>>> Good point.
> >>>>
> >>>> But set_config() should be rare:
> >>>>
> >>>> E.g in the case of virtio-net with VERSION_1, config space is read only,
> >>>> and it was set via control vq.
> >>>>
> >>>> For block, we can
> >>>>
> >>>> 1) start from without WCE or
> >>>> 2) we add a config change notification to userspace or
> >>> I prefer this way. And I think we also need to do similar things for
> >>> set/get_vq_state().
> >>
> >> Yes, I agree.
> >>
> > Hi Jason,
> >
> > Now I'm working on this. But I found the config change notification
> > must be synchronous in the virtio-blk case, which means the kernel
> > still needs to wait for the response from userspace in set_config().
> > Otherwise, some I/Os might still run the old way after we change the
> > cache_type in sysfs.
> >
> > The simple ways to solve this problem are:
> >
> > 1. Only support read-only config space, disable WCE as you suggested
> > 2. Add a return value to set_config() and handle the failure only in
> > virtio-blk driver
> > 3. Print some warnings after timeout since it only affects the
> > dataplane which is under userspace's control
> >
> > Any suggestions?
>
>
> Let's go without WCE first and make VDUSE work first. We can then think
> of a solution for WCE on top.
>

It's fine with me.

Thanks,
Yongji

Powered by blists - more mailing lists