[<prev] [next>] [day] [month] [year] [list]
Message-ID: <5c2c5fb8-f85a-a3e3-24b4-2103a8ddaff7@intel.com>
Date: Fri, 27 Jul 2018 09:16:28 +0800
From: Jie Deng <jie.deng@...el.com>
To: "Chen, Jian Jun" <jian.jun.chen@...el.com>,
"kraxel@...hat.com" <kraxel@...hat.com>
Cc: jie.deng@...el.com, "mst@...hat.com" <mst@...hat.com>,
"jasowang@...hat.com" <jasowang@...hat.com>,
"Wang, Yu1" <yu1.wang@...el.com>,
"Liu, Shuo A" <shuo.a.liu@...el.com>,
"Chen, Conghui" <conghui.chen@...el.com>,
virtualization@...ts.linux-foundation.org,
linux-kernel@...r.kernel.org
Subject: Re: why vdev->config->reset(vdev) is not called in virtinput_freeze
Hi Gerd,
Do you have any comments ?
On 2018/7/23 16:56, Chen, Jian Jun wrote:
> Hi Gerd Hoffmann,
>
> vdev->config->reset(vdev) is not called before vdev->config->del_vqs(vdev) in virtinput_freeze.
> Backend driver cannot receive notification about the suspend.
> Is it possible that backend driver tries to write a buffer pointed by a desc which has been freed by frontend driver?
>
> Can you please explain why vdev->config->reset(vdev) is not called in virtinput_freeze?
> Virtio_net/virtio_block/virtio_rng... all have the reset in their pm suspend callback.
>
> Best regards,
> Chen Jian Jun
>
Regards,
Jie
Powered by blists - more mailing lists