1NVME-FORMAT(1) NVMe Manual NVME-FORMAT(1)
2
6 nvme-format - Format an NVMe device
7
9 nvme format <device> [--namespace-id=<nsid> | -n <nsid>]
10 [--lbaf=<lbaf> | -l <lbaf>]
11 [--block-size=<block size | -b <block size>]
12 [--ses=<ses> | -s <ses>]
13 [--pil=<pil> | -p <pil>]
14 [--pi=<pi> | -i <pi>]
15 [--ms=<ms> | -m <ms>]
16 [--reset | -r ]
17 [--timeout=<timeout> | -t <timeout> ]
18
20 For the NVMe device given, send an nvme Format Namespace admin command
21 and provides the results.
22 The <device> parameter is mandatory and may be either the NVMe
24 character device (ex: /dev/nvme0), or a namespace block device (ex:
25 /dev/nvme0n1). If the character device is given, the namespace
26 identifier will default to 0xffffffff to send the format to all
27 namespace, but can be overridden to any namespace with the namespace-id
28 option. If the block device is given, the namespace identifier will
29 default to the namespace id of the block device given, but can be
30 overridden with the same option.
31 On success, the program will automatically issue BLKRRPART ioctl to
33 force rescanning the namespaces. If the driver is recent enough, this
34 will automatically update the physical block size. If it is not recent
35 enough, you will need to remove and rescan your device some other way
36 for the new block size to be visible.
37
39 -n <nsid>, --namespace-id=<nsid>
40 Send the format command for the specified nsid. This can be used to
41 override the default value for either character device (0xffffffff)
42 or the block device (result from NVME_IOCTL_ID).
43 -l <lbaf>, --lbaf=<lbaf>
45 LBA Format: This field specifies the LBA format to apply to the NVM
46 media. This corresponds to the LBA formats indicated in the
47 Identify Namespace command. Conflicts with --block-size argument.
48 Defaults to 0.
49 -b <block size>, --block-size=<block size>
51 Block Size: This field is used to specify the target block size to
52 format to. Potential lbaf values will be scanned and the lowest
53 numbered will be selected for the format operation. Conflicts with
54 --lbaf argument.
55 -s <ses>, --ses=<ses>
57 Secure Erase Settings: This field specifies whether a secure erase
58 should be performed as part of the format and the type of the
59 secure erase operation. The erase applies to all user data,
60 regardless of location (e.g., within an exposed LBA, within a
61 cache, within deallocated LBAs, etc). Defaults to 0.
62 ┌──────┬────────────────────────────┐
64 │Value │ Definition │
65 ├──────┼────────────────────────────┤
66 │0 │ No secure erase operation │
67 │ │ requested │
68 ├──────┼────────────────────────────┤
69 │1 │ User Data Erase: All user │
70 │ │ data shall be erased, │
71 │ │ contents of the user data │
72 │ │ after the erase is │
73 │ │ indeterminate (e.g., the │
74 │ │ user data may be zero │
75 │ │ filled, one filled, etc). │
76 │ │ The controller may perform │
77 │ │ a cryptographic erase when │
78 │ │ a User Data Erase is │
79 │ │ requested if all user data │
80 │ │ is encrypted. │
81 ├──────┼────────────────────────────┤
82 │2 │ Cryptographic Erase: All │
83 │ │ user data shall be erased │
84 │ │ cryptographically. This is │
85 │ │ accomplished by deleting │
86 │ │ the encryption key. │
87 ├──────┼────────────────────────────┤
88 │3–7 │ Reserved │
89 └──────┴────────────────────────────┘
90 -p <pil>, --pil=<pil>
92 Protection Information Location: If set to ‘1’ and protection
93 information is enabled, then protection information is transferred
94 as the first eight bytes of metadata. If cleared to ‘0’ and
95 protection information is enabled, then protection information is
96 transferred as the last eight bytes of metadata. Defaults to 0.
97 -i <pi>, --pi=<pi>
99 Protection Information: This field specifies whether end-to-end
100 data protection is enabled and the type of protection information.
101 Defaults to 0.
102 ┌──────┬───────────────────────────┐
104 │Value │ Definition │
105 ├──────┼───────────────────────────┤
106 │0 │ Protection information is │
107 │ │ not enabled │
108 ├──────┼───────────────────────────┤
109 │1 │ Protection information is │
110 │ │ enabled, Type 1 │
111 ├──────┼───────────────────────────┤
112 │2 │ Protection information is │
113 │ │ enabled, Type 2 │
114 ├──────┼───────────────────────────┤
115 │3 │ Protection information is │
116 │ │ enabled, Type 3 │
117 ├──────┼───────────────────────────┤
118 │4–7 │ Reserved │
119 └──────┴───────────────────────────┘
120 -m <ms>, --ms=<ms>
122 Metadata Settings: This field is set to ‘1’ if the metadata is
123 transferred as part of an extended data LBA. This field is cleared
124 to ‘0’ if the metadata is transferred as part of a separate buffer.
125 The metadata may include protection information, based on the
126 Protection Information (PI) field. Defaults to 0.
127 -r, --reset
129 Issue a reset after successful format. Must use the character
130 device for this.
131 -t <timeout>, --timeout=<timeout>
133 Override default timeout value. In milliseconds.
134
136 · Format the device using all defaults:
137 # nvme format /dev/nvme0n1
139 · Format namespace 1 with user data secure erase settings and
141 protection information:
142 # nvme format /dev/nvme0 --namespace-id=1 --ses=1 --pi=1
144
146 Part of the nvme-user suite
147NVMe 01/08/2019 NVME-FORMAT(1)