docs/specs/qcow2.txt

   1 == General ==
   2
   3 A qcow2 image file is organized in units of constant size, which are called
   4 (host) clusters. A cluster is the unit in which all allocations are done,
   5 both for actual guest data and for image metadata.
   6
   7 Likewise, the virtual disk as seen by the guest is divided into (guest)
   8 clusters of the same size.
   9
  10 All numbers in qcow2 are stored in Big Endian byte order.
  11
  12
  13 == Header ==
  14
  15 The first cluster of a qcow2 image contains the file header:
  16
  17     Byte  0 -  3:   magic
  18                     QCOW magic string ("QFI\xfb")
  19
  20           4 -  7:   version
  21                     Version number (valid values are 2 and 3)
  22
  23           8 - 15:   backing_file_offset
  24                     Offset into the image file at which the backing file name
  25                     is stored (NB: The string is not null terminated). 0 if the
  26                     image doesn't have a backing file.
  27
  28          16 - 19:   backing_file_size
  29                     Length of the backing file name in bytes. Must not be
  30                     longer than 1023 bytes. Undefined if the image doesn't have
  31                     a backing file.
  32
  33          20 - 23:   cluster_bits
  34                     Number of bits that are used for addressing an offset
  35                     within a cluster (1 << cluster_bits is the cluster size).
  36                     Must not be less than 9 (i.e. 512 byte clusters).
  37
  38                     Note: qemu as of today has an implementation limit of 2 MB
  39                     as the maximum cluster size and won't be able to open images
  40                     with larger cluster sizes.
  41
  42          24 - 31:   size
  43                     Virtual disk size in bytes
  44
  45          32 - 35:   crypt_method
  46                     0 for no encryption
  47                     1 for AES encryption
  48
  49          36 - 39:   l1_size
  50                     Number of entries in the active L1 table
  51
  52          40 - 47:   l1_table_offset
  53                     Offset into the image file at which the active L1 table
  54                     starts. Must be aligned to a cluster boundary.
  55
  56          48 - 55:   refcount_table_offset
  57                     Offset into the image file at which the refcount table
  58                     starts. Must be aligned to a cluster boundary.
  59
  60          56 - 59:   refcount_table_clusters
  61                     Number of clusters that the refcount table occupies
  62
  63          60 - 63:   nb_snapshots
  64                     Number of snapshots contained in the image
  65
  66          64 - 71:   snapshots_offset
  67                     Offset into the image file at which the snapshot table
  68                     starts. Must be aligned to a cluster boundary.
  69
  70 If the version is 3 or higher, the header has the following additional fields.
  71 For version 2, the values are assumed to be zero, unless specified otherwise
  72 in the description of a field.
  73
  74          72 -  79:  incompatible_features
  75                     Bitmask of incompatible features. An implementation must
  76                     fail to open an image if an unknown bit is set.
  77
  78                     Bits 0-63:  Reserved (set to 0)
  79
  80          80 -  87:  compatible_features
  81                     Bitmask of compatible features. An implementation can
  82                     safely ignore any unknown bits that are set.
  83
  84                     Bits 0-63:  Reserved (set to 0)
  85
  86          88 -  95:  autoclear_features
  87                     Bitmask of auto-clear features. An implementation may only
  88                     write to an image with unknown auto-clear features if it
  89                     clears the respective bits from this field first.
  90
  91                     Bits 0-63:  Reserved (set to 0)
  92
  93          96 -  99:  refcount_order
  94                     Describes the width of a reference count block entry (width
  95                     in bits = 1 << refcount_order). For version 2 images, the
  96                     order is always assumed to be 4 (i.e. the width is 16 bits).
  97
  98         100 - 103:  header_length
  99                     Length of the header structure in bytes. For version 2
 100                     images, the length is always assumed to be 72 bytes.
 101
 102 Directly after the image header, optional sections called header extensions can
 103 be stored. Each extension has a structure like the following:
 104
 105     Byte  0 -  3:   Header extension type:
 106                         0x00000000 - End of the header extension area
 107                         0xE2792ACA - Backing file format name
 108                         0x6803f857 - Feature name table
 109                         other      - Unknown header extension, can be safely
 110                                      ignored
 111
 112           4 -  7:   Length of the header extension data
 113
 114           8 -  n:   Header extension data
 115
 116           n -  m:   Padding to round up the header extension size to the next
 117                     multiple of 8.
 118
 119 Unless stated otherwise, each header extension type shall appear at most once
 120 in the same image.
 121
 122 The remaining space between the end of the header extension area and the end of
 123 the first cluster can be used for the backing file name. It is not allowed to
 124 store other data here, so that an implementation can safely modify the header
 125 and add extensions without harming data of compatible features that it
 126 doesn't support. Compatible features that need space for additional data can
 127 use a header extension.
 128
 129
 130 == Feature name table ==
 131
 132 The feature name table is an optional header extension that contains the name
 133 for features used by the image. It can be used by applications that don't know
 134 the respective feature (e.g. because the feature was introduced only later) to
 135 display a useful error message.
 136
 137 The number of entries in the feature name table is determined by the length of
 138 the header extension data. Each entry look like this:
 139
 140     Byte       0:   Type of feature (select feature bitmap)
 141                         0: Incompatible feature
 142                         1: Compatible feature
 143                         2: Autoclear feature
 144
 145                1:   Bit number within the selected feature bitmap (valid
 146                     values: 0-63)
 147
 148           2 - 47:   Feature name (padded with zeros, but not necessarily null
 149                     terminated if it has full length)
 150
 151
 152 == Host cluster management ==
 153
 154 qcow2 manages the allocation of host clusters by maintaining a reference count
 155 for each host cluster. A refcount of 0 means that the cluster is free, 1 means
 156 that it is used, and >= 2 means that it is used and any write access must
 157 perform a COW (copy on write) operation.
 158
 159 The refcounts are managed in a two-level table. The first level is called
 160 refcount table and has a variable size (which is stored in the header). The
 161 refcount table can cover multiple clusters, however it needs to be contiguous
 162 in the image file.
 163
 164 It contains pointers to the second level structures which are called refcount
 165 blocks and are exactly one cluster in size.
 166
 167 Given a offset into the image file, the refcount of its cluster can be obtained
 168 as follows:
 169
 170     refcount_block_entries = (cluster_size / sizeof(uint16_t))
 171
 172     refcount_block_index = (offset / cluster_size) % refcount_block_entries
 173     refcount_table_index = (offset / cluster_size) / refcount_block_entries
 174
 175     refcount_block = load_cluster(refcount_table[refcount_table_index]);
 176     return refcount_block[refcount_block_index];
 177
 178 Refcount table entry:
 179
 180     Bit  0 -  8:    Reserved (set to 0)
 181
 182          9 - 63:    Bits 9-63 of the offset into the image file at which the
 183                     refcount block starts. Must be aligned to a cluster
 184                     boundary.
 185
 186                     If this is 0, the corresponding refcount block has not yet
 187                     been allocated. All refcounts managed by this refcount block
 188                     are 0.
 189
 190 Refcount block entry (x = refcount_bits - 1):
 191
 192     Bit  0 -  x:    Reference count of the cluster. If refcount_bits implies a
 193                     sub-byte width, note that bit 0 means the least significant
 194                     bit in this context.
 195
 196
 197 == Cluster mapping ==
 198
 199 Just as for refcounts, qcow2 uses a two-level structure for the mapping of
 200 guest clusters to host clusters. They are called L1 and L2 table.
 201
 202 The L1 table has a variable size (stored in the header) and may use multiple
 203 clusters, however it must be contiguous in the image file. L2 tables are
 204 exactly one cluster in size.
 205
 206 Given a offset into the virtual disk, the offset into the image file can be
 207 obtained as follows:
 208
 209     l2_entries = (cluster_size / sizeof(uint64_t))
 210
 211     l2_index = (offset / cluster_size) % l2_entries
 212     l1_index = (offset / cluster_size) / l2_entries
 213
 214     l2_table = load_cluster(l1_table[l1_index]);
 215     cluster_offset = l2_table[l2_index];
 216
 217     return cluster_offset + (offset % cluster_size)
 218
 219 L1 table entry:
 220
 221     Bit  0 -  8:    Reserved (set to 0)
 222
 223          9 - 55:    Bits 9-55 of the offset into the image file at which the L2
 224                     table starts. Must be aligned to a cluster boundary. If the
 225                     offset is 0, the L2 table and all clusters described by this
 226                     L2 table are unallocated.
 227
 228         56 - 62:    Reserved (set to 0)
 229
 230              63:    0 for an L2 table that is unused or requires COW, 1 if its
 231                     refcount is exactly one. This information is only accurate
 232                     in the active L1 table.
 233
 234 L2 table entry:
 235
 236     Bit  0 -  61:   Cluster descriptor
 237
 238               62:   0 for standard clusters
 239                     1 for compressed clusters
 240
 241               63:   0 for a cluster that is unused or requires COW, 1 if its
 242                     refcount is exactly one. This information is only accurate
 243                     in L2 tables that are reachable from the the active L1
 244                     table.
 245
 246 Standard Cluster Descriptor:
 247
 248     Bit       0:    If set to 1, the cluster reads as all zeros. The host
 249                     cluster offset can be used to describe a preallocation,
 250                     but it won't be used for reading data from this cluster,
 251                     nor is data read from the backing file if the cluster is
 252                     unallocated.
 253
 254                     With version 2, this is always 0.
 255
 256          1 -  8:    Reserved (set to 0)
 257
 258          9 - 55:    Bits 9-55 of host cluster offset. Must be aligned to a
 259                     cluster boundary. If the offset is 0, the cluster is
 260                     unallocated.
 261
 262         56 - 61:    Reserved (set to 0)
 263
 264
 265 Compressed Clusters Descriptor (x = 62 - (cluster_size - 8)):
 266
 267     Bit  0 -  x:    Host cluster offset. This is usually _not_ aligned to a
 268                     cluster boundary!
 269
 270        x+1 - 61:    Compressed size of the images in sectors of 512 bytes
 271
 272 If a cluster is unallocated, read requests shall read the data from the backing
 273 file (except if bit 0 in the Standard Cluster Descriptor is set). If there is
 274 no backing file or the backing file is smaller than the image, they shall read
 275 zeros for all parts that are not covered by the backing file.
 276
 277
 278 == Snapshots ==
 279
 280 qcow2 supports internal snapshots. Their basic principle of operation is to
 281 switch the active L1 table, so that a different set of host clusters are
 282 exposed to the guest.
 283
 284 When creating a snapshot, the L1 table should be copied and the refcount of all
 285 L2 tables and clusters reachable from this L1 table must be increased, so that
 286 a write causes a COW and isn't visible in other snapshots.
 287
 288 When loading a snapshot, bit 63 of all entries in the new active L1 table and
 289 all L2 tables referenced by it must be reconstructed from the refcount table
 290 as it doesn't need to be accurate in inactive L1 tables.
 291
 292 A directory of all snapshots is stored in the snapshot table, a contiguous area
 293 in the image file, whose starting offset and length are given by the header
 294 fields snapshots_offset and nb_snapshots. The entries of the snapshot table
 295 have variable length, depending on the length of ID, name and extra data.
 296
 297 Snapshot table entry:
 298
 299     Byte 0 -  7:    Offset into the image file at which the L1 table for the
 300                     snapshot starts. Must be aligned to a cluster boundary.
 301
 302          8 - 11:    Number of entries in the L1 table of the snapshots
 303
 304         12 - 13:    Length of the unique ID string describing the snapshot
 305
 306         14 - 15:    Length of the name of the snapshot
 307
 308         16 - 19:    Time at which the snapshot was taken in seconds since the
 309                     Epoch
 310
 311         20 - 23:    Subsecond part of the time at which the snapshot was taken
 312                     in nanoseconds
 313
 314         24 - 31:    Time that the guest was running until the snapshot was
 315                     taken in nanoseconds
 316
 317         32 - 35:    Size of the VM state in bytes. 0 if no VM state is saved.
 318                     If there is VM state, it starts at the first cluster
 319                     described by first L1 table entry that doesn't describe a
 320                     regular guest cluster (i.e. VM state is stored like guest
 321                     disk content, except that it is stored at offsets that are
 322                     larger than the virtual disk presented to the guest)
 323
 324         36 - 39:    Size of extra data in the table entry (used for future
 325                     extensions of the format)
 326
 327         variable:   Extra data for future extensions. Unknown fields must be
 328                     ignored. Currently defined are (offset relative to snapshot
 329                     table entry):
 330
 331                     Byte 40 - 47:   Size of the VM state in bytes. 0 if no VM
 332                                     state is saved. If this field is present,
 333                                     the 32-bit value in bytes 32-35 is ignored.
 334
 335                     Byte 48 - 55:   Virtual disk size of the snapshot in bytes
 336
 337                     Version 3 images must include extra data at least up to
 338                     byte 55.
 339
 340         variable:   Unique ID string for the snapshot (not null terminated)
 341
 342         variable:   Name of the snapshot (not null terminated)