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 Bits 1-63: Reserved (set to 0)
85 80 - 87: compatible_features
86 Bitmask of compatible features. An implementation can
87 safely ignore any unknown bits that are set.
89 Bit 0: Lazy refcounts bit. If this bit is set then
90 lazy refcount updates can be used. This means
91 marking the image file dirty and postponing
92 refcount metadata updates.
94 Bits 1-63: Reserved (set to 0)
96 88 - 95: autoclear_features
97 Bitmask of auto-clear features. An implementation may only
98 write to an image with unknown auto-clear features if it
99 clears the respective bits from this field first.
101 Bits 0-63: Reserved (set to 0)
103 96 - 99: refcount_order
104 Describes the width of a reference count block entry (width
105 in bits = 1 << refcount_order). For version 2 images, the
106 order is always assumed to be 4 (i.e. the width is 16 bits).
108 100 - 103: header_length
109 Length of the header structure in bytes. For version 2
110 images, the length is always assumed to be 72 bytes.
112 Directly after the image header, optional sections called header extensions can
113 be stored. Each extension has a structure like the following:
115 Byte 0 - 3: Header extension type:
116 0x00000000 - End of the header extension area
117 0xE2792ACA - Backing file format name
118 0x6803f857 - Feature name table
119 other - Unknown header extension, can be safely
122 4 - 7: Length of the header extension data
124 8 - n: Header extension data
126 n - m: Padding to round up the header extension size to the next
129 Unless stated otherwise, each header extension type shall appear at most once
132 The remaining space between the end of the header extension area and the end of
133 the first cluster can be used for the backing file name. It is not allowed to
134 store other data here, so that an implementation can safely modify the header
135 and add extensions without harming data of compatible features that it
136 doesn't support. Compatible features that need space for additional data can
137 use a header extension.
140 == Feature name table ==
142 The feature name table is an optional header extension that contains the name
143 for features used by the image. It can be used by applications that don't know
144 the respective feature (e.g. because the feature was introduced only later) to
145 display a useful error message.
147 The number of entries in the feature name table is determined by the length of
148 the header extension data. Each entry look like this:
150 Byte 0: Type of feature (select feature bitmap)
151 0: Incompatible feature
152 1: Compatible feature
155 1: Bit number within the selected feature bitmap (valid
158 2 - 47: Feature name (padded with zeros, but not necessarily null
159 terminated if it has full length)
162 == Host cluster management ==
164 qcow2 manages the allocation of host clusters by maintaining a reference count
165 for each host cluster. A refcount of 0 means that the cluster is free, 1 means
166 that it is used, and >= 2 means that it is used and any write access must
167 perform a COW (copy on write) operation.
169 The refcounts are managed in a two-level table. The first level is called
170 refcount table and has a variable size (which is stored in the header). The
171 refcount table can cover multiple clusters, however it needs to be contiguous
174 It contains pointers to the second level structures which are called refcount
175 blocks and are exactly one cluster in size.
177 Given a offset into the image file, the refcount of its cluster can be obtained
180 refcount_block_entries = (cluster_size / sizeof(uint16_t))
182 refcount_block_index = (offset / cluster_size) % refcount_block_entries
183 refcount_table_index = (offset / cluster_size) / refcount_block_entries
185 refcount_block = load_cluster(refcount_table[refcount_table_index]);
186 return refcount_block[refcount_block_index];
188 Refcount table entry:
190 Bit 0 - 8: Reserved (set to 0)
192 9 - 63: Bits 9-63 of the offset into the image file at which the
193 refcount block starts. Must be aligned to a cluster
196 If this is 0, the corresponding refcount block has not yet
197 been allocated. All refcounts managed by this refcount block
200 Refcount block entry (x = refcount_bits - 1):
202 Bit 0 - x: Reference count of the cluster. If refcount_bits implies a
203 sub-byte width, note that bit 0 means the least significant
207 == Cluster mapping ==
209 Just as for refcounts, qcow2 uses a two-level structure for the mapping of
210 guest clusters to host clusters. They are called L1 and L2 table.
212 The L1 table has a variable size (stored in the header) and may use multiple
213 clusters, however it must be contiguous in the image file. L2 tables are
214 exactly one cluster in size.
216 Given a offset into the virtual disk, the offset into the image file can be
219 l2_entries = (cluster_size / sizeof(uint64_t))
221 l2_index = (offset / cluster_size) % l2_entries
222 l1_index = (offset / cluster_size) / l2_entries
224 l2_table = load_cluster(l1_table[l1_index]);
225 cluster_offset = l2_table[l2_index];
227 return cluster_offset + (offset % cluster_size)
231 Bit 0 - 8: Reserved (set to 0)
233 9 - 55: Bits 9-55 of the offset into the image file at which the L2
234 table starts. Must be aligned to a cluster boundary. If the
235 offset is 0, the L2 table and all clusters described by this
236 L2 table are unallocated.
238 56 - 62: Reserved (set to 0)
240 63: 0 for an L2 table that is unused or requires COW, 1 if its
241 refcount is exactly one. This information is only accurate
242 in the active L1 table.
246 Bit 0 - 61: Cluster descriptor
248 62: 0 for standard clusters
249 1 for compressed clusters
251 63: 0 for a cluster that is unused or requires COW, 1 if its
252 refcount is exactly one. This information is only accurate
253 in L2 tables that are reachable from the the active L1
256 Standard Cluster Descriptor:
258 Bit 0: If set to 1, the cluster reads as all zeros. The host
259 cluster offset can be used to describe a preallocation,
260 but it won't be used for reading data from this cluster,
261 nor is data read from the backing file if the cluster is
264 With version 2, this is always 0.
266 1 - 8: Reserved (set to 0)
268 9 - 55: Bits 9-55 of host cluster offset. Must be aligned to a
269 cluster boundary. If the offset is 0, the cluster is
272 56 - 61: Reserved (set to 0)
275 Compressed Clusters Descriptor (x = 62 - (cluster_bits - 8)):
277 Bit 0 - x: Host cluster offset. This is usually _not_ aligned to a
280 x+1 - 61: Compressed size of the images in sectors of 512 bytes
282 If a cluster is unallocated, read requests shall read the data from the backing
283 file (except if bit 0 in the Standard Cluster Descriptor is set). If there is
284 no backing file or the backing file is smaller than the image, they shall read
285 zeros for all parts that are not covered by the backing file.
290 qcow2 supports internal snapshots. Their basic principle of operation is to
291 switch the active L1 table, so that a different set of host clusters are
292 exposed to the guest.
294 When creating a snapshot, the L1 table should be copied and the refcount of all
295 L2 tables and clusters reachable from this L1 table must be increased, so that
296 a write causes a COW and isn't visible in other snapshots.
298 When loading a snapshot, bit 63 of all entries in the new active L1 table and
299 all L2 tables referenced by it must be reconstructed from the refcount table
300 as it doesn't need to be accurate in inactive L1 tables.
302 A directory of all snapshots is stored in the snapshot table, a contiguous area
303 in the image file, whose starting offset and length are given by the header
304 fields snapshots_offset and nb_snapshots. The entries of the snapshot table
305 have variable length, depending on the length of ID, name and extra data.
307 Snapshot table entry:
309 Byte 0 - 7: Offset into the image file at which the L1 table for the
310 snapshot starts. Must be aligned to a cluster boundary.
312 8 - 11: Number of entries in the L1 table of the snapshots
314 12 - 13: Length of the unique ID string describing the snapshot
316 14 - 15: Length of the name of the snapshot
318 16 - 19: Time at which the snapshot was taken in seconds since the
321 20 - 23: Subsecond part of the time at which the snapshot was taken
324 24 - 31: Time that the guest was running until the snapshot was
327 32 - 35: Size of the VM state in bytes. 0 if no VM state is saved.
328 If there is VM state, it starts at the first cluster
329 described by first L1 table entry that doesn't describe a
330 regular guest cluster (i.e. VM state is stored like guest
331 disk content, except that it is stored at offsets that are
332 larger than the virtual disk presented to the guest)
334 36 - 39: Size of extra data in the table entry (used for future
335 extensions of the format)
337 variable: Extra data for future extensions. Unknown fields must be
338 ignored. Currently defined are (offset relative to snapshot
341 Byte 40 - 47: Size of the VM state in bytes. 0 if no VM
342 state is saved. If this field is present,
343 the 32-bit value in bytes 32-35 is ignored.
345 Byte 48 - 55: Virtual disk size of the snapshot in bytes
347 Version 3 images must include extra data at least up to
350 variable: Unique ID string for the snapshot (not null terminated)
352 variable: Name of the snapshot (not null terminated)