5 QEMU provides NVMe emulation through the ``nvme``, ``nvme-ns`` and
6 ``nvme-subsys`` devices.
8 See the following sections for specific information on
10 * `Adding NVMe Devices`_, `additional namespaces`_ and `NVM subsystems`_.
11 * Configuration of `Optional Features`_ such as `Controller Memory Buffer`_,
12 `Simple Copy`_, `Zoned Namespaces`_, `metadata`_ and `End-to-End Data
21 The QEMU emulated NVMe controller implements version 1.4 of the NVM Express
22 specification. All mandatory features are implement with a couple of exceptions
25 * Accounting numbers in the SMART/Health log page are reset when the device
27 * Interrupt Coalescing is not supported and is disabled by default.
29 The simplest way to attach an NVMe controller on the QEMU PCI bus is to add the
32 .. code-block:: console
34 -drive file=nvm.img,if=none,id=nvm
35 -device nvme,serial=deadbeef,drive=nvm
37 There are a number of optional general parameters for the ``nvme`` device. Some
38 are mentioned here, but see ``-device nvme,help`` to list all possible
41 ``max_ioqpairs=UINT32`` (default: ``64``)
42 Set the maximum number of allowed I/O queue pairs. This replaces the
43 deprecated ``num_queues`` parameter.
45 ``msix_qsize=UINT16`` (default: ``65``)
46 The number of MSI-X vectors that the device should support.
48 ``mdts=UINT8`` (default: ``7``)
49 Set the Maximum Data Transfer Size of the device.
51 ``use-intel-id`` (default: ``off``)
52 Since QEMU 5.2, the device uses a QEMU allocated "Red Hat" PCI Device and
53 Vendor ID. Set this to ``on`` to revert to the unallocated Intel ID
59 In the simplest possible invocation sketched above, the device only support a
60 single namespace with the namespace identifier ``1``. To support multiple
61 namespaces and additional features, the ``nvme-ns`` device must be used.
63 .. code-block:: console
65 -device nvme,id=nvme-ctrl-0,serial=deadbeef
66 -drive file=nvm-1.img,if=none,id=nvm-1
67 -device nvme-ns,drive=nvm-1
68 -drive file=nvm-2.img,if=none,id=nvm-2
69 -device nvme-ns,drive=nvm-2
71 The namespaces defined by the ``nvme-ns`` device will attach to the most
72 recently defined ``nvme-bus`` that is created by the ``nvme`` device. Namespace
73 identifiers are allocated automatically, starting from ``1``.
75 There are a number of parameters available:
77 ``nsid`` (default: ``0``)
78 Explicitly set the namespace identifier.
80 ``uuid`` (default: *autogenerated*)
81 Set the UUID of the namespace. This will be reported as a "Namespace UUID"
82 descriptor in the Namespace Identification Descriptor List.
85 Set the EUI-64 of the namespace. This will be reported as a "IEEE Extended
86 Unique Identifier" descriptor in the Namespace Identification Descriptor List.
87 Since machine type 6.1 a non-zero default value is used if the parameter
88 is not provided. For earlier machine types the field defaults to 0.
91 If there are more ``nvme`` devices defined, this parameter may be used to
92 attach the namespace to a specific ``nvme`` device (identified by an ``id``
93 parameter on the controller device).
98 Additional features becomes available if the controller device (``nvme``) is
99 linked to an NVM Subsystem device (``nvme-subsys``).
101 The NVM Subsystem emulation allows features such as shared namespaces and
104 .. code-block:: console
106 -device nvme-subsys,id=nvme-subsys-0,nqn=subsys0
107 -device nvme,serial=deadbeef,subsys=nvme-subsys-0
108 -device nvme,serial=deadbeef,subsys=nvme-subsys-0
110 This will create an NVM subsystem with two controllers. Having controllers
111 linked to an ``nvme-subsys`` device allows additional ``nvme-ns`` parameters:
113 ``shared`` (default: ``on`` since 6.2)
114 Specifies that the namespace will be attached to all controllers in the
115 subsystem. If set to ``off``, the namespace will remain a private namespace
116 and may only be attached to a single controller at a time. Shared namespaces
117 are always automatically attached to all controllers (also when controllers
120 ``detached`` (default: ``off``)
121 If set to ``on``, the namespace will be be available in the subsystem, but
122 not attached to any controllers initially. A shared namespace with this set
123 to ``on`` will never be automatically attached to controllers.
127 .. code-block:: console
129 -drive file=nvm-1.img,if=none,id=nvm-1
130 -device nvme-ns,drive=nvm-1,nsid=1
131 -drive file=nvm-2.img,if=none,id=nvm-2
132 -device nvme-ns,drive=nvm-2,nsid=3,shared=off,detached=on
134 will cause NSID 1 will be a shared namespace that is initially attached to both
135 controllers. NSID 3 will be a private namespace due to ``shared=off`` and only
136 attachable to a single controller at a time. Additionally it will not be
137 attached to any controller initially (due to ``detached=on``) or to hotplugged
143 Controller Memory Buffer
144 ------------------------
146 ``nvme`` device parameters related to the Controller Memory Buffer support:
148 ``cmb_size_mb=UINT32`` (default: ``0``)
149 This adds a Controller Memory Buffer of the given size at offset zero in BAR
152 ``legacy-cmb`` (default: ``off``)
153 By default, the device uses the "v1.4 scheme" for the Controller Memory
154 Buffer support (i.e, the CMB is initially disabled and must be explicitly
155 enabled by the host). Set this to ``on`` to behave as a v1.3 device wrt. the
161 The device includes support for TP 4065 ("Simple Copy Command"). A number of
162 additional ``nvme-ns`` device parameters may be used to control the Copy
165 ``mssrl=UINT16`` (default: ``128``)
166 Set the Maximum Single Source Range Length (``MSSRL``). This is the maximum
167 number of logical blocks that may be specified in each source range.
169 ``mcl=UINT32`` (default: ``128``)
170 Set the Maximum Copy Length (``MCL``). This is the maximum number of logical
171 blocks that may be specified in a Copy command (the total for all source
174 ``msrc=UINT8`` (default: ``127``)
175 Set the Maximum Source Range Count (``MSRC``). This is the maximum number of
176 source ranges that may be used in a Copy command. This is a 0's based value.
181 A namespaces may be "Zoned" as defined by TP 4053 ("Zoned Namespaces"). Set
182 ``zoned=on`` on an ``nvme-ns`` device to configure it as a zoned namespace.
184 The namespace may be configured with additional parameters
186 ``zoned.zone_size=SIZE`` (default: ``128MiB``)
187 Define the zone size (``ZSZE``).
189 ``zoned.zone_capacity=SIZE`` (default: ``0``)
190 Define the zone capacity (``ZCAP``). If left at the default (``0``), the zone
191 capacity will equal the zone size.
193 ``zoned.descr_ext_size=UINT32`` (default: ``0``)
194 Set the Zone Descriptor Extension Size (``ZDES``). Must be a multiple of 64
197 ``zoned.cross_read=BOOL`` (default: ``off``)
198 Set to ``on`` to allow reads to cross zone boundaries.
200 ``zoned.max_active=UINT32`` (default: ``0``)
201 Set the maximum number of active resources (``MAR``). The default (``0``)
202 allows all zones to be active.
204 ``zoned.max_open=UINT32`` (default: ``0``)
205 Set the maximum number of open resources (``MOR``). The default (``0``)
206 allows all zones to be open. If ``zoned.max_active`` is specified, this value
207 must be less than or equal to that.
209 ``zoned.zasl=UINT8`` (default: ``0``)
210 Set the maximum data transfer size for the Zone Append command. Like
211 ``mdts``, the value is specified as a power of two (2^n) and is in units of
212 the minimum memory page size (CAP.MPSMIN). The default value (``0``)
213 has this property inherit the ``mdts`` value.
218 The virtual namespace device supports LBA metadata in the form separate
219 metadata (``MPTR``-based) and extended LBAs.
221 ``ms=UINT16`` (default: ``0``)
222 Defines the number of metadata bytes per LBA.
224 ``mset=UINT8`` (default: ``0``)
225 Set to ``1`` to enable extended LBAs.
227 End-to-End Data Protection
228 --------------------------
230 The virtual namespace device supports DIF- and DIX-based protection information
231 (depending on ``mset``).
233 ``pi=UINT8`` (default: ``0``)
234 Enable protection information of the specified type (type ``1``, ``2`` or
237 ``pil=UINT8`` (default: ``0``)
238 Controls the location of the protection information within the metadata. Set
239 to ``1`` to transfer protection information as the first eight bytes of
240 metadata. Otherwise, the protection information is transferred as the last
243 Virtualization Enhancements and SR-IOV (Experimental Support)
244 -------------------------------------------------------------
246 The ``nvme`` device supports Single Root I/O Virtualization and Sharing
247 along with Virtualization Enhancements. The controller has to be linked to
248 an NVM Subsystem device (``nvme-subsys``) for use with SR-IOV.
250 A number of parameters are present (**please note, that they may be
251 subject to change**):
253 ``sriov_max_vfs`` (default: ``0``)
254 Indicates the maximum number of PCIe virtual functions supported
255 by the controller. Specifying a non-zero value enables reporting of both
256 SR-IOV and ARI (Alternative Routing-ID Interpretation) capabilities
257 by the NVMe device. Virtual function controllers will not report SR-IOV.
259 ``sriov_vq_flexible``
260 Indicates the total number of flexible queue resources assignable to all
261 the secondary controllers. Implicitly sets the number of primary
262 controller's private resources to ``(max_ioqpairs - sriov_vq_flexible)``.
264 ``sriov_vi_flexible``
265 Indicates the total number of flexible interrupt resources assignable to
266 all the secondary controllers. Implicitly sets the number of primary
267 controller's private resources to ``(msix_qsize - sriov_vi_flexible)``.
269 ``sriov_max_vi_per_vf`` (default: ``0``)
270 Indicates the maximum number of virtual interrupt resources assignable
271 to a secondary controller. The default ``0`` resolves to
272 ``(sriov_vi_flexible / sriov_max_vfs)``
274 ``sriov_max_vq_per_vf`` (default: ``0``)
275 Indicates the maximum number of virtual queue resources assignable to
276 a secondary controller. The default ``0`` resolves to
277 ``(sriov_vq_flexible / sriov_max_vfs)``
279 The simplest possible invocation enables the capability to set up one VF
280 controller and assign an admin queue, an IO queue, and a MSI-X interrupt.
282 .. code-block:: console
284 -device nvme-subsys,id=subsys0
285 -device nvme,serial=deadbeef,subsys=subsys0,sriov_max_vfs=1,
286 sriov_vq_flexible=2,sriov_vi_flexible=1
288 The minimum steps required to configure a functional NVMe secondary
291 * unbind flexible resources from the primary controller
293 .. code-block:: console
295 nvme virt-mgmt /dev/nvme0 -c 0 -r 1 -a 1 -n 0
296 nvme virt-mgmt /dev/nvme0 -c 0 -r 0 -a 1 -n 0
298 * perform a Function Level Reset on the primary controller to actually
299 release the resources
301 .. code-block:: console
303 echo 1 > /sys/bus/pci/devices/0000:01:00.0/reset
307 .. code-block:: console
309 echo 1 > /sys/bus/pci/devices/0000:01:00.0/sriov_numvfs
311 * assign the flexible resources to the VF and set it ONLINE
313 .. code-block:: console
315 nvme virt-mgmt /dev/nvme0 -c 1 -r 1 -a 8 -n 1
316 nvme virt-mgmt /dev/nvme0 -c 1 -r 0 -a 8 -n 2
317 nvme virt-mgmt /dev/nvme0 -c 1 -r 0 -a 9 -n 0
319 * bind the NVMe driver to the VF
321 .. code-block:: console
323 echo 0000:01:00.1 > /sys/bus/pci/drivers/nvme/bind