δΊ 2013-2-28 0:00, Markus Armbruster ει:
> Humor me: no period at end of subject, please
>
OK.
> First pass, concentrating on interfaces, implementation mostly ignored.
>
> Wenchao Xia <xiaw...@linux.vnet.ibm.com> writes:
>
>> This mirror function will return all image info including
>> snapshots, if specified backing image's info will also be returned.
>> Now Qemu have both query-images and query-block interfaces.
>>
>> Signed-off-by: Wenchao Xia <xiaw...@linux.vnet.ibm.com>
>> ---
>> block.c | 83
>> ++++++++++++++++++++++++++++++++++++++++++++++++++++++
>> qapi-schema.json | 34 ++++++++++++++++++++++
>> qmp-commands.hx | 76 +++++++++++++++++++++++++++++++++++++++++++++++++
>> 3 files changed, 193 insertions(+), 0 deletions(-)
>>
> [...]
>> diff --git a/qapi-schema.json b/qapi-schema.json
>> index cd7ea25..9122bb9 100644
>> --- a/qapi-schema.json
>> +++ b/qapi-schema.json
>> @@ -245,6 +245,22 @@
>> '*backing-filename-format': 'str', '*snapshots':
>> ['SnapshotInfo'] } }
>>
>> ##
>> +# @DeviceImageInfo:
>> +#
>> +# Information about an image used by a QEMU block device
>> +#
>> +# @device: name of the block device
>> +#
>> +# @image: #optional info of the image used
>> +#
>> +# Since: 1.5
>> +#
>> +##
>> +
>> +{ 'type': 'DeviceImageInfo',
>> + 'data': {'device': 'str', '*image': 'ImageInfo' } }
>> +
>> +##
>> # @StatusInfo:
>> #
>> # Information about VCPU run state
>> @@ -789,6 +805,24 @@
>> { 'command': 'query-block', 'returns': ['BlockInfo'] }
>>
>> ##
>> +# @query-images:
>> +#
>> +# Get a list of DeviceImageInfo for all virtual block devices.
>
> # Get block device image information
>
OK.
>> +#
>> +# @device: #optional the name of the device to get image info from. If not
>> +# specified, all block devices will be queried
>> +# @backing: #optional true to show information on backing images, false or
>> +# omitted to show just the top image of a block device
>
> I'm not sure these flags are necessary.
>
These flag seems more flex to me. Always querying all info require
caller do a filter operation on what he got, but in many times caller
may be interested on only one device.
>> +#
>> +# Returns: a list of @DeviceImageInfo describing each virtual block device
>> +#
>> +# Since: 1.5
>> +##
>> +{ 'command': 'query-images',
>> + 'data': { '*device': 'str', '*backing': 'bool' },
>> + 'returns': ['DeviceImageInfo'] }
>> +
>> +##
>> # @BlockDeviceStats:
>> #
>> # Statistics of a virtual block device or a block backing device.
>> diff --git a/qmp-commands.hx b/qmp-commands.hx
>> index 799adea..292d61e 100644
>> --- a/qmp-commands.hx
>> +++ b/qmp-commands.hx
>> @@ -1743,6 +1743,82 @@ EQMP
>> },
>>
>> SQMP
>> +query-images
>> +-----------
>> +
>> +Show the block devices' images.
>
> Show block device image information
>
OK.
>> +
>> +Each block image information is stored in a json-object and the returned
>> value
>> +is a json-array of all devices' images.
>
> Each device is represented by a json-object. The returned value is a
> json-array
> of all devices.
>
OK.
>> +
>> +Each json-object contain the following:
>> +
>> +- "device": device name (json-string)
>> +- "image": related image information, it is a optional json-object
>> containing the following:
>
> - "image": image information (json-object, optional) containing:
>
OK.
>> + - "filename": image file name (json-string)
>> + - "format": image format (json-string)
>> + - "virtual-size": maximum capacity in bytes of the image (json-int)
>
> - "virtual-size": image capacity in bytes (json-int)
>
>> + - "dirty-flag": true if image is not cleanly closed (json-bool,
>> optional)
>
> Long line.
>
> What's the meaning of dirty-flag not present in object?
>
>> + - "actual-size": actual size on disk in bytes of the image
>> (json-int, optional)
>
> - "actual-size": actual on-disk image size in bytes
> (json-int, optional)
>
> Can you explain when actual-size is present, and when not? Same for the
> other optional members.
>
OK, will document them.
>> + - "cluster-size": size of a cluster in bytes (json-int, optional)
>> + - "encrypted": true if the image is encrypted (json-bool, optional)
>> + - "backing_file": backing file name (json-string, optional)
>> + - "full-backing-filename": full path of the backing file
>> (json-string, optional)
>> + - "backing-filename-format": the format of the backing file
>> (json-string, optional)
>
> Long lines.
>
>> + - "snapshots": the internal snapshot info, it is an optional list
>> of json-object
>> + containing the following:
>
> - "snapshots": internal snapshot info
> (json-array of json-object, optional)
> Each object contains:
>
>> + - "id": unique snapshot id (json-string)
>> + - "name": internal snapshot name (json-string)
>
> Suggest to drop "internal"
>
>> + - "vm-state-size": size of the VM state in bytes (json-int)
>> + - "date-sec": UTC date of the snapshot in seconds (json-int)
>> + - "date-nsec": fractional part in nano seconds to be used with
>> date-sec(json-int)
>> + - "vm-clock-sec": VM clock relative to boot in seconds
>> (json-int)
>> + - "vm-clock-nsec": fractional part in nano seconds to be used
>> with vm-clock-sec (json-int)
>
> Long lines.
>
> A year has 86400 s == 8.64e13 ns, thus int64_t is good for more than
> 10000 years. Why can't we simply measure in ns and dispense with the
> awkward (sec, nsec) pair?
>
I am not sure this brings much benefits. In most case the management
stack may need to use the "seconds" to show for end user.
>> +
>> +Example:
>> +
>> +-> { "execute": "query-images" }
>> +<- {
>> + "return":[
>> + {
>> + "device":"ide0-hd0",
>> + "image":{
>> + "filename":"disks/test0.img",
>> + "format":"qcow2",
>> + "virtual-size":1024000
>> + }
>> + },
>> + {
>> + "device":"ide0-hd1",
>> + "image":{
>> + "filename":"disks/test1.img",
>> + "format":"qcow2",
>> + "virtual-size":2048000,
>> + "snapshots":[
>> + {
>> + "id": "1",
>> + "name": "snapshot1",
>> + "vm-state-size": 0,
>> + "date-sec": 10000200,
>> + "date-nsec": 12,
>> + "vm-clock-sec": 206,
>> + "vm-clock-nsec": 30
>> + }
>> + ]
>> + }
>> + }
>> + ]
>> + }
>> +
>> +EQMP
>> +
>> + {
>> + .name = "query-images",
>> + .args_type = "device:B?, backing:-b",
>> + .mhandler.cmd_new = qmp_marshal_input_query_images,
>> + },
>> +
>> +SQMP
>> query-blockstats
>> ----------------
>
--
Best Regards
Wenchao Xia
