Re: [Qemu-devel] [PATCH v5 RFC] spec: add qcow2 bitmaps extension specification

[Vladimir Sementsov-Ogievskiy <vsementsov@virtuozzo.com>]

On 22.12.2015 02:56, Max Reitz wrote:
> On 21.12.2015 16:25, Vladimir Sementsov-Ogievskiy wrote:
>> The new feature for qcow2: storing bitmaps.
>>
>> Only bitmaps, relative to the virtual disk, stored in qcow2 file, should
>> be stored in this qcow2 file.
>>
>> Strings started from +# are RFC-strings, not to be commited of course
>>
>> Signed-off-by: Vladimir Sementsov-Ogievskiy <vsementsov@virtuozzo.com>
>> ---
>>
>>
>> v5:
>>
>> - 'Dirty bitmaps' renamed to 'Bitmaps', as we may have several types of
>>    bitmaps.
>> - rewordings
>> - move upper bounds to "Notes about Qemu limits"
>> - s/should/must somewhere. (but not everywhere)
>> - move name_size field closer to name itself in bitmap header
>> - add extra data area to bitmap header
>> - move bitmap data description to separate section
>>
>>
>>
>>   docs/specs/qcow2.txt | 160 ++++++++++++++++++++++++++++++++++++++++++++++++++-
>>   1 file changed, 159 insertions(+), 1 deletion(-)
> Looks good! :-)

Thank you! )

>
> Some comments below, but I think the general design is good now.
>
>> diff --git a/docs/specs/qcow2.txt b/docs/specs/qcow2.txt
>> index 121dfc8..3d557ee 100644
>> --- a/docs/specs/qcow2.txt
>> +++ b/docs/specs/qcow2.txt
>> @@ -103,7 +103,19 @@ in the description of a field.
>>                       write to an image with unknown auto-clear features if it
>>                       clears the respective bits from this field first.
>>   
>> -                    Bits 0-63:  Reserved (set to 0)
>> +                    Bit 0:      Bitmaps extension bit.
>> +                                This bit is responsible for Bitmaps extension
>> +                                consistency.
>> +
>> +                                If it is set, but there is no Bitmaps
>> +                                extension, this should be considered as an
>> +                                error.
>> +
>> +                                If it is not set, but there is a Bitmaps
>> +                                extension, its data should be considered as
>> +                                inconsistent.
>> +
>> +                    Bits 1-63:  Reserved (set to 0)
>>   
>>            96 -  99:  refcount_order
>>                       Describes the width of a reference count block entry (width
>> @@ -123,6 +135,7 @@ be stored. Each extension has a structure like the following:
>>                           0x00000000 - End of the header extension area
>>                           0xE2792ACA - Backing file format name
>>                           0x6803f857 - Feature name table
>> +                        0x23852875 - Bitmaps extension
>>                           other      - Unknown header extension, can be safely
>>                                        ignored
>>   
>> @@ -166,6 +179,34 @@ the header extension data. Each entry look like this:
>>                       terminated if it has full length)
>>   
>>   
>> +== Bitmaps extension ==
>> +
>> +Bitmaps extension is an optional header extension. It provides an ability to
>> +store virtual disk related bitmaps in a qcow2 image. For now there is only one
>> +type of such bitmaps: Dirty Tracking Bitmap, which just tracks virtual disk
>> +changes from some moment.
>> +
>> +The data of the extension should be considered as consistent only if
>> +corresponding auto-clear feature bit is set (see autoclear_features above).
>> +
>> +The fields of Bitmaps extension are:
>> +
>> +          0 -  3:  nb_bitmaps
>> +                   The number of bitmaps contained in the image. Must be
>> +                   greater or equal to 1.
>> +
>> +                   Note: Qemu currently only supports up to 65535 bitmaps per
>> +                   image.
>> +
>> +          4 -  7:  bitmap_directory_size
>> +                   Size of the Bitmap Directory in bytes. It must be equal to
>> +                   sum of sizes of all (nb_bitmaps) bitmap headers.
> I'd rather write this as: "Size of the Bitmap Directory in bytes, i.e.
> the cumulative size of all (nb_bitmaps) bitmap headers."
>
> ("It must" sounds like it's an additional restriction while it's
> actually just an explanation.)
>
>> +
>> +          8 - 15:  bitmap_directory_offset
>> +                   Offset into the image file at which the Bitmap Directory
>> +                   starts. Must be aligned to a cluster boundary.
>> +
>> +
>>   == Host cluster management ==
>>   
>>   qcow2 manages the allocation of host clusters by maintaining a reference count
>> @@ -360,3 +401,120 @@ Snapshot table entry:
>>   
>>           variable:   Padding to round up the snapshot table entry size to the
>>                       next multiple of 8.
>> +
>> +
>> +== Bitmaps ==
>> +
>> +The feature supports storing bitmaps in a qcow2 image. All bitmaps are related
>> +to the virtual disk, stored in this image.
>> +
>> +=== Bitmap Directory ===
>> +
>> +Each bitmap saved in the image is described in a Bitmap Directory entry. Bitmap
>> +Directory is a contiguous area in the image file, whose starting offset and
>> +length are given by the header extension fields bitmap_directory_offset and
>> +bitmap_directory_size. The entries of the bitmap directory have variable
>> +length, depending on the length of the bitmap name and extra data. These
>> +entries are also called bitmap headers.
>> +
>> +Bitmap Directory Entry:
>> +
>> +    Byte 0 -  7:    bitmap_table_offset
>> +                    Offset into the image file at which the Bitmap Table
>> +                    (described below) for the bitmap starts. Must be aligned to
>> +                    a cluster boundary.
>> +
>> +         8 - 11:    bitmap_table_size
>> +                    Number of entries in the Bitmap Table of the bitmap.
>> +
>> +        12 - 15:    flags
>> +                    Bit
>> +                      0: in_use
>> +                         The bitmap was not saved correctly and may be
>> +                         inconsistent.
>> +
>> +                      1: auto
>> +                         The bitmap must reflect all changes of the virtual
>> +                         disk by any application that would write to this qcow2
>> +                         file (including writes, snapshot switching, etc.). The
>> +                         type of this bitmap must be 'Dirty Tracking Bitmap'.
>> +
>> +                    Bits 2 - 31 are reserved and must be 0.
>> +
>> +             16:    type
>> +                    This field describes the sort of the bitmap.
>> +                    Values:
>> +                      0: Dirty Tracking Bitmap
>> +
>> +                    Values 1 - 255 are reserved.
>> +# Is there any reason, to make it '1 - Dirty Tracking Bitmap, 0,2-255 - reserved?'
> Maybe something arbitrary like "But 0 feels like 'unknown' to me!". But
> I can't think of anything substantial.
>
>> +
>> +             17:    granularity_bits
>> +                    Granularity bits. Valid values are: 0 - 63.
>> +
>> +                    Note: Qemu currently doesn't support granularity_bits
>> +                    greater than 31.
>> +
>> +                    Granularity is calculated as
>> +                        granularity = 1 << granularity_bits
>> +
>> +                    Granularity of the bitmap is how many bytes of the image
>> +                    accounts for one bit of the bitmap.
>> +
>> +        18 - 19:    name_size
>> +                    Size of the bitmap name. Valid values: 1 - 1023.
>> +
>> +        20 - 21:    extra_data_size
>> +                    Size of type-specific extra data.
>> +# 2 bytes or 4 bytes ?
> I'd make it 4. I don't think those additional two bytes will hurt
> anybody, and the simple fact that you're asking that question shows to
> me that there is a possibility that 2 won't be enough. ;-)
>
>> +
>> +        variable:   The name of the bitmap (not null terminated). Must be
>> +                    unique among all bitmap names within the Bitmaps extension.
>> +
>> +        variable:   Type-specific extra data for the bitmap.
> I'd swap name and type-specific data so we know the alignment of the
> type-specific data (if they are swapped and sizeof(extra_data_size) ==
> 4, its alignment will be 4). Maybe that'll come in handy at some point.

Good thing, thanks. It will be even 8)

>
> Maybe you should make a note that currently, no extra data is defined
> for Dirty Tracking Bitmaps, but that this may change in the future.
>
> Max
>
>> +        variable:   Padding to round up the Bitmap Directory Entry size to the
>> +                    next multiple of 8.
>> +
>> +=== Bitmap Table ===
>> +
>> +Bitmaps are stored using a one-level (not two-level like refcounts and guest
>> +clusters mapping) structure for the mapping of bitmaps data to host clusters.
>> +It is called Bitmap Table.
>> +
>> +Each Bitmap Table has a variable size (stored in the Bitmap Directory Entry)
>> +and may use multiple clusters, however it must be contiguous in the image file.
>> +
>> +Bitmap Table entry:
>> +
>> +    Bit       0:    Reserved and must be zero if bits 9 - 55 are non-zero.
>> +                    If bits 9 - 55 are zero:
>> +                      0: Cluster should be read as all zeros.
>> +                      1: Cluster should be read as all ones.
>> +
>> +         1 -  8:    Reserved and must be zero.
>> +
>> +         9 - 55:    Bits 9 - 55 of host cluster offset. Must be aligned to a
>> +                    cluster boundary. If the offset is 0, the cluster is
>> +                    unallocated, see bit 0 description.
>> +
>> +        56 - 63:    Reserved and must be zero.
>> +
>> +=== Bitmap Data ===
>> +
>> +As noted above, bitmap data is stored in several (or may be one, exactly
>> +bitmap_table_size) separate clusters, described by Bitmap Table. Given an
>> +offset (in bytes) into the bitmap data, the offset into the image file can be
>> +obtained as follows:
>> +
>> +    image_offset =
>> +        bitmap_table[bitmap_data_offset / cluster_size] +
>> +            (bitmap_data_offset % cluster_size)
>> +
>> +Taking into account the granularity of the bitmap, an offset in bits into the
>> +image file, corresponding to byte number byte_nr of the virtual disk can be
>> +calculated like this:
>> +
>> +    bit_offset =
>> +        image_offset(byte_nr / granularity / 8) * 8 +
>> +            (byte_nr / granularity) % 8
>>
>


-- 
Best regards,
Vladimir
* now, @virtuozzo.com instead of @parallels.com. Sorry for this inconvenience.