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.
7 Likewise, the virtual disk as seen by the guest is divided into (guest)
8 clusters of the same size.
10 All numbers in qcow2 are stored in Big Endian byte order.
15 The first cluster of a qcow2 image contains the file header:
18 QCOW magic string ("QFI\xfb")
21 Version number (valid values are 2 and 3)
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.
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
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).
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.
43 Virtual disk size in bytes
50 Number of entries in the active L1 table
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.
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.
60 56 - 59: refcount_table_clusters
61 Number of clusters that the refcount table occupies
64 Number of snapshots contained in the image
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.
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.
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.
78 Bit 0: Dirty bit. If this bit is set then refcounts
79 may be inconsistent, make sure to scan L1/L2
80 tables to repair refcounts before accessing the
83 Bit 1: Corrupt bit. If this bit is set then any data
84 structure may be corrupt and the image must not
85 be written to (unless for regaining
88 Bits 2-63: Reserved (set to 0)
90 80 - 87: compatible_features
91 Bitmask of compatible features. An implementation can
92 safely ignore any unknown bits that are set.
94 Bit 0: Lazy refcounts bit. If this bit is set then
95 lazy refcount updates can be used. This means
96 marking the image file dirty and postponing
97 refcount metadata updates.
99 Bits 1-63: Reserved (set to 0)
101 88 - 95: autoclear_features
102 Bitmask of auto-clear features. An implementation may only
103 write to an image with unknown auto-clear features if it
104 clears the respective bits from this field first.
106 Bits 0-63: Reserved (set to 0)
108 96 - 99: refcount_order
109 Describes the width of a reference count block entry (width
110 in bits: refcount_bits = 1 << refcount_order). For version 2
111 images, the order is always assumed to be 4
112 (i.e. refcount_bits = 16).
114 100 - 103: header_length
115 Length of the header structure in bytes. For version 2
116 images, the length is always assumed to be 72 bytes.
118 Directly after the image header, optional sections called header extensions can
119 be stored. Each extension has a structure like the following:
121 Byte 0 - 3: Header extension type:
122 0x00000000 - End of the header extension area
123 0xE2792ACA - Backing file format name
124 0x6803f857 - Feature name table
125 other - Unknown header extension, can be safely
128 4 - 7: Length of the header extension data
130 8 - n: Header extension data
132 n - m: Padding to round up the header extension size to the next
135 Unless stated otherwise, each header extension type shall appear at most once
138 The remaining space between the end of the header extension area and the end of
139 the first cluster can be used for the backing file name. It is not allowed to
140 store other data here, so that an implementation can safely modify the header
141 and add extensions without harming data of compatible features that it
142 doesn't support. Compatible features that need space for additional data can
143 use a header extension.
146 == Feature name table ==
148 The feature name table is an optional header extension that contains the name
149 for features used by the image. It can be used by applications that don't know
150 the respective feature (e.g. because the feature was introduced only later) to
151 display a useful error message.
153 The number of entries in the feature name table is determined by the length of
154 the header extension data. Each entry look like this:
156 Byte 0: Type of feature (select feature bitmap)
157 0: Incompatible feature
158 1: Compatible feature
161 1: Bit number within the selected feature bitmap (valid
164 2 - 47: Feature name (padded with zeros, but not necessarily null
165 terminated if it has full length)
168 == Host cluster management ==
170 qcow2 manages the allocation of host clusters by maintaining a reference count
171 for each host cluster. A refcount of 0 means that the cluster is free, 1 means
172 that it is used, and >= 2 means that it is used and any write access must
173 perform a COW (copy on write) operation.
175 The refcounts are managed in a two-level table. The first level is called
176 refcount table and has a variable size (which is stored in the header). The
177 refcount table can cover multiple clusters, however it needs to be contiguous
180 It contains pointers to the second level structures which are called refcount
181 blocks and are exactly one cluster in size.
183 Given a offset into the image file, the refcount of its cluster can be obtained
186 refcount_block_entries = (cluster_size / sizeof(uint16_t))
188 refcount_block_index = (offset / cluster_size) % refcount_block_entries
189 refcount_table_index = (offset / cluster_size) / refcount_block_entries
191 refcount_block = load_cluster(refcount_table[refcount_table_index]);
192 return refcount_block[refcount_block_index];
194 Refcount table entry:
196 Bit 0 - 8: Reserved (set to 0)
198 9 - 63: Bits 9-63 of the offset into the image file at which the
199 refcount block starts. Must be aligned to a cluster
202 If this is 0, the corresponding refcount block has not yet
203 been allocated. All refcounts managed by this refcount block
206 Refcount block entry (x = refcount_bits - 1):
208 Bit 0 - x: Reference count of the cluster. If refcount_bits implies a
209 sub-byte width, note that bit 0 means the least significant
213 == Cluster mapping ==
215 Just as for refcounts, qcow2 uses a two-level structure for the mapping of
216 guest clusters to host clusters. They are called L1 and L2 table.
218 The L1 table has a variable size (stored in the header) and may use multiple
219 clusters, however it must be contiguous in the image file. L2 tables are
220 exactly one cluster in size.
222 Given a offset into the virtual disk, the offset into the image file can be
225 l2_entries = (cluster_size / sizeof(uint64_t))
227 l2_index = (offset / cluster_size) % l2_entries
228 l1_index = (offset / cluster_size) / l2_entries
230 l2_table = load_cluster(l1_table[l1_index]);
231 cluster_offset = l2_table[l2_index];
233 return cluster_offset + (offset % cluster_size)
237 Bit 0 - 8: Reserved (set to 0)
239 9 - 55: Bits 9-55 of the offset into the image file at which the L2
240 table starts. Must be aligned to a cluster boundary. If the
241 offset is 0, the L2 table and all clusters described by this
242 L2 table are unallocated.
244 56 - 62: Reserved (set to 0)
246 63: 0 for an L2 table that is unused or requires COW, 1 if its
247 refcount is exactly one. This information is only accurate
248 in the active L1 table.
252 Bit 0 - 61: Cluster descriptor
254 62: 0 for standard clusters
255 1 for compressed clusters
257 63: 0 for a cluster that is unused or requires COW, 1 if its
258 refcount is exactly one. This information is only accurate
259 in L2 tables that are reachable from the the active L1
262 Standard Cluster Descriptor:
264 Bit 0: If set to 1, the cluster reads as all zeros. The host
265 cluster offset can be used to describe a preallocation,
266 but it won't be used for reading data from this cluster,
267 nor is data read from the backing file if the cluster is
270 With version 2, this is always 0.
272 1 - 8: Reserved (set to 0)
274 9 - 55: Bits 9-55 of host cluster offset. Must be aligned to a
275 cluster boundary. If the offset is 0, the cluster is
278 56 - 61: Reserved (set to 0)
281 Compressed Clusters Descriptor (x = 62 - (cluster_bits - 8)):
283 Bit 0 - x: Host cluster offset. This is usually _not_ aligned to a
286 x+1 - 61: Compressed size of the images in sectors of 512 bytes
288 If a cluster is unallocated, read requests shall read the data from the backing
289 file (except if bit 0 in the Standard Cluster Descriptor is set). If there is
290 no backing file or the backing file is smaller than the image, they shall read
291 zeros for all parts that are not covered by the backing file.
296 qcow2 supports internal snapshots. Their basic principle of operation is to
297 switch the active L1 table, so that a different set of host clusters are
298 exposed to the guest.
300 When creating a snapshot, the L1 table should be copied and the refcount of all
301 L2 tables and clusters reachable from this L1 table must be increased, so that
302 a write causes a COW and isn't visible in other snapshots.
304 When loading a snapshot, bit 63 of all entries in the new active L1 table and
305 all L2 tables referenced by it must be reconstructed from the refcount table
306 as it doesn't need to be accurate in inactive L1 tables.
308 A directory of all snapshots is stored in the snapshot table, a contiguous area
309 in the image file, whose starting offset and length are given by the header
310 fields snapshots_offset and nb_snapshots. The entries of the snapshot table
311 have variable length, depending on the length of ID, name and extra data.
313 Snapshot table entry:
315 Byte 0 - 7: Offset into the image file at which the L1 table for the
316 snapshot starts. Must be aligned to a cluster boundary.
318 8 - 11: Number of entries in the L1 table of the snapshots
320 12 - 13: Length of the unique ID string describing the snapshot
322 14 - 15: Length of the name of the snapshot
324 16 - 19: Time at which the snapshot was taken in seconds since the
327 20 - 23: Subsecond part of the time at which the snapshot was taken
330 24 - 31: Time that the guest was running until the snapshot was
333 32 - 35: Size of the VM state in bytes. 0 if no VM state is saved.
334 If there is VM state, it starts at the first cluster
335 described by first L1 table entry that doesn't describe a
336 regular guest cluster (i.e. VM state is stored like guest
337 disk content, except that it is stored at offsets that are
338 larger than the virtual disk presented to the guest)
340 36 - 39: Size of extra data in the table entry (used for future
341 extensions of the format)
343 variable: Extra data for future extensions. Unknown fields must be
344 ignored. Currently defined are (offset relative to snapshot
347 Byte 40 - 47: Size of the VM state in bytes. 0 if no VM
348 state is saved. If this field is present,
349 the 32-bit value in bytes 32-35 is ignored.
351 Byte 48 - 55: Virtual disk size of the snapshot in bytes
353 Version 3 images must include extra data at least up to
356 variable: Unique ID string for the snapshot (not null terminated)
358 variable: Name of the snapshot (not null terminated)
360 variable: Padding to round up the snapshot table entry size to the