[CRIU] [PATCH v3 09/16] fdinfo: ARTIFICIAL fd type introduced

Pavel Emelyanov xemul at parallels.com
Mon Dec 14 04:02:42 PST 2015


On 12/14/2015 02:57 PM, Stanislav Kinsburskiy wrote:
> 
> 
> 14.12.2015 12:48, Pavel Emelyanov пишет:
>> On 12/14/2015 02:17 PM, Stanislav Kinsburskiy wrote:
>>>
>>> 14.12.2015 12:10, Pavel Emelyanov пишет:
>>>> On 12/10/2015 06:16 PM, Stanislav Kinsburskiy wrote:
>>>>> Will be used to distinguish between service descriptors and noreman ones.
>>>>>
>>>>> Signed-off-by: Stanislav Kinsburskiy <skinsbursky at virtuozzo.com>
>>>>> ---
>>>>>    protobuf/fdinfo.proto |    1 +
>>>>>    1 file changed, 1 insertion(+)
>>>>>
>>>>> diff --git a/protobuf/fdinfo.proto b/protobuf/fdinfo.proto
>>>>> index ac80f62..eec8cd6 100644
>>>>> --- a/protobuf/fdinfo.proto
>>>>> +++ b/protobuf/fdinfo.proto
>>>>> @@ -17,6 +17,7 @@ enum fd_types {
>>>>>    	TUNF		= 15;
>>>>>    	EXT		= 16;
>>>>>    	TIMERFD		= 17;
>>>>> +	ARTIFICIAL	= 18;
>>>> This type never gets into images, so there should be no such constant in
>>>> images description.
>>> The intention was to highlight such files in restore log to distinguish
>>> between other files and them.
>>> I can drop them, but what should be the replacement?
>> Some constant but not in the protobuf file?
> 
> You mean some variable, representing read_fd?
> Or some boolean constant in addition to read_fd, representing, whether 
> read_fd is valid?
> What does "not in the protobuf file" means? Some other file?

We're talking here about the FD_TYPES__ARTIFICIAL constant only. This constant
should not be declared in the image description as it never gets there.

-- Pavel



More information about the CRIU mailing list