[<prev] [next>] [day] [month] [year] [list]
Message-ID: <b34df002-344c-f011-dce0-fee882f62af2@redhat.com>
Date: Sun, 26 Apr 2020 15:38:42 +0800
From: Jason Wang <jasowang@...hat.com>
To: Zhu Lingshan <lingshan.zhu@...ux.intel.com>,
Zhu Lingshan <lingshan.zhu@...el.com>, mst@...hat.com,
kvm@...r.kernel.org, virtualization@...ts.linux-foundation.org,
linux-kernel@...r.kernel.org, netdev@...r.kernel.org
Cc: lulu@...hat.com, dan.daly@...el.com, cunming.liang@...el.com
Subject: Re: [PATCH V2 1/2] vdpa: Support config interrupt in vhost_vdpa
On 2020/4/26 下午3:24, Zhu Lingshan wrote:
>
>
> On 4/26/2020 3:03 PM, Jason Wang wrote:
>>
>> On 2020/4/26 下午2:58, Jason Wang wrote:
>>>>
>>>> diff --git a/drivers/vhost/vhost.h b/drivers/vhost/vhost.h
>>>> index 1813821..8663139 100644
>>>> --- a/drivers/vhost/vhost.h
>>>> +++ b/drivers/vhost/vhost.h
>>>> @@ -18,6 +18,8 @@
>>>> typedef void (*vhost_work_fn_t)(struct vhost_work *work);
>>>> #define VHOST_WORK_QUEUED 1
>>>> +#define VHOST_FILE_UNBIND -1
>>>
>>>
>>> I think it's better to document this in uapi.
>>
>>
>> I meant e.g in vhost_vring_file, we had a comment of unbinding:
>>
>> struct vhost_vring_file {
>> unsigned int index;
>> int fd; /* Pass -1 to unbind from file. */
>>
>> };
> I think it is better to use an int fd than vhost_vring_file, to avoid the confusions,
> so if we add#define VHOST_FILE_UNBIND -1 in the uapi header, can it document itself?
> Thanks
I think so.
Thanks
>>
>> Thanks
>>
Powered by blists - more mailing lists