]>
Commit | Line | Data |
---|---|---|
6e4c46c4 DC |
1 | [[qm_pci_passthrough]] |
2 | PCI(e) Passthrough | |
3 | ------------------ | |
e582833b DC |
4 | ifdef::wiki[] |
5 | :pve-toplevel: | |
6 | endif::wiki[] | |
6e4c46c4 DC |
7 | |
8 | PCI(e) passthrough is a mechanism to give a virtual machine control over | |
49f20f1b TL |
9 | a PCI device from the host. This can have some advantages over using |
10 | virtualized hardware, for example lower latency, higher performance, or more | |
11 | features (e.g., offloading). | |
6e4c46c4 | 12 | |
49f20f1b | 13 | But, if you pass through a device to a virtual machine, you cannot use that |
6e4c46c4 DC |
14 | device anymore on the host or in any other VM. |
15 | ||
16 | General Requirements | |
17 | ~~~~~~~~~~~~~~~~~~~~ | |
18 | ||
19 | Since passthrough is a feature which also needs hardware support, there are | |
49f20f1b TL |
20 | some requirements to check and preparations to be done to make it work. |
21 | ||
6e4c46c4 DC |
22 | |
23 | Hardware | |
24 | ^^^^^^^^ | |
49f20f1b TL |
25 | Your hardware needs to support `IOMMU` (*I*/*O* **M**emory **M**anagement |
26 | **U**nit) interrupt remapping, this includes the CPU and the mainboard. | |
6e4c46c4 | 27 | |
49f20f1b TL |
28 | Generally, Intel systems with VT-d, and AMD systems with AMD-Vi support this. |
29 | But it is not guaranteed that everything will work out of the box, due | |
30 | to bad hardware implementation and missing or low quality drivers. | |
6e4c46c4 | 31 | |
49f20f1b | 32 | Further, server grade hardware has often better support than consumer grade |
6e4c46c4 DC |
33 | hardware, but even then, many modern system can support this. |
34 | ||
49f20f1b TL |
35 | Please refer to your hardware vendor to check if they support this feature |
36 | under Linux for your specific setup | |
37 | ||
6e4c46c4 DC |
38 | |
39 | Configuration | |
40 | ^^^^^^^^^^^^^ | |
41 | ||
49f20f1b TL |
42 | Once you ensured that your hardware supports passthrough, you will need to do |
43 | some configuration to enable PCI(e) passthrough. | |
6e4c46c4 | 44 | |
6e4c46c4 | 45 | |
39d84f28 | 46 | .IOMMU |
6e4c46c4 | 47 | |
49f20f1b | 48 | The IOMMU has to be activated on the kernel commandline. The easiest way is to |
39d84f28 | 49 | enable trough grub. Edit `'/etc/default/grub'' and add the following to the |
49f20f1b | 50 | 'GRUB_CMDLINE_LINUX_DEFAULT' variable: |
6e4c46c4 | 51 | |
49f20f1b TL |
52 | * for Intel CPUs: |
53 | + | |
54 | ---- | |
55 | intel_iommu=on | |
56 | ---- | |
57 | * for AMD CPUs: | |
58 | + | |
59 | ---- | |
6e4c46c4 | 60 | amd_iommu=on |
49f20f1b | 61 | ---- |
6e4c46c4 | 62 | |
39d84f28 | 63 | [[qm_pci_passthrough_update_grub]] |
49f20f1b TL |
64 | To bring this change in effect, make sure you run: |
65 | ||
66 | ---- | |
67 | # update-grub | |
68 | ---- | |
6e4c46c4 | 69 | |
39d84f28 | 70 | .Kernel Modules |
6e4c46c4 | 71 | |
49f20f1b TL |
72 | You have to make sure the following modules are loaded. This can be achieved by |
73 | adding them to `'/etc/modules'' | |
6e4c46c4 | 74 | |
49f20f1b | 75 | ---- |
6e4c46c4 DC |
76 | vfio |
77 | vfio_iommu_type1 | |
78 | vfio_pci | |
79 | vfio_virqfd | |
49f20f1b | 80 | ---- |
6e4c46c4 | 81 | |
49f20f1b | 82 | [[qm_pci_passthrough_update_initramfs]] |
6e4c46c4 | 83 | After changing anything modules related, you need to refresh your |
49f20f1b | 84 | `initramfs`. On {pve} this can be done by executing: |
6e4c46c4 DC |
85 | |
86 | ---- | |
49f20f1b | 87 | # update-initramfs -u -k all |
6e4c46c4 DC |
88 | ---- |
89 | ||
39d84f28 | 90 | .Finish Configuration |
49f20f1b TL |
91 | |
92 | Finally reboot to bring the changes into effect and check that it is indeed | |
93 | enabled. | |
6e4c46c4 DC |
94 | |
95 | ---- | |
5e235b99 | 96 | # dmesg | grep -e DMAR -e IOMMU -e AMD-Vi |
6e4c46c4 DC |
97 | ---- |
98 | ||
49f20f1b TL |
99 | should display that `IOMMU`, `Directed I/O` or `Interrupt Remapping` is |
100 | enabled, depending on hardware and kernel the exact message can vary. | |
6e4c46c4 DC |
101 | |
102 | It is also important that the device(s) you want to pass through | |
49f20f1b | 103 | are in a *separate* `IOMMU` group. This can be checked with: |
6e4c46c4 DC |
104 | |
105 | ---- | |
49f20f1b | 106 | # find /sys/kernel/iommu_groups/ -type l |
6e4c46c4 DC |
107 | ---- |
108 | ||
49f20f1b | 109 | It is okay if the device is in an `IOMMU` group together with its functions |
6e4c46c4 DC |
110 | (e.g. a GPU with the HDMI Audio device) or with its root port or PCI(e) bridge. |
111 | ||
112 | .PCI(e) slots | |
113 | [NOTE] | |
114 | ==== | |
49f20f1b TL |
115 | Some platforms handle their physical PCI(e) slots differently. So, sometimes |
116 | it can help to put the card in a another PCI(e) slot, if you do not get the | |
117 | desired `IOMMU` group separation. | |
6e4c46c4 DC |
118 | ==== |
119 | ||
120 | .Unsafe interrupts | |
121 | [NOTE] | |
122 | ==== | |
123 | For some platforms, it may be necessary to allow unsafe interrupts. | |
49f20f1b TL |
124 | For this add the following line in a file ending with `.conf' file in |
125 | */etc/modprobe.d/*: | |
6e4c46c4 | 126 | |
49f20f1b | 127 | ---- |
6e4c46c4 | 128 | options vfio_iommu_type1 allow_unsafe_interrupts=1 |
49f20f1b | 129 | ---- |
6e4c46c4 DC |
130 | |
131 | Please be aware that this option can make your system unstable. | |
132 | ==== | |
133 | ||
082b32fb TL |
134 | GPU Passthrough Notes |
135 | ^^^^^^^^^^^^^^^^^^^^^ | |
13cae0c1 | 136 | |
082b32fb TL |
137 | It is not possible to display the frame buffer of the GPU via NoVNC or SPICE on |
138 | the {pve} web interface. | |
13cae0c1 | 139 | |
082b32fb TL |
140 | When passing through a whole GPU or a vGPU and graphic output is wanted, one |
141 | has to either physically connect a monitor to the card, or configure a remote | |
142 | desktop software (for example, VNC or RDP) inside the guest. | |
13cae0c1 | 143 | |
082b32fb TL |
144 | If you want to use the GPU as a hardware accelerator, for example, for |
145 | programs using OpenCL or CUDA, this is not required. | |
13cae0c1 | 146 | |
49f20f1b | 147 | Host Device Passthrough |
6e4c46c4 DC |
148 | ~~~~~~~~~~~~~~~~~~~~~~~ |
149 | ||
150 | The most used variant of PCI(e) passthrough is to pass through a whole | |
49f20f1b TL |
151 | PCI(e) card, for example a GPU or a network card. |
152 | ||
6e4c46c4 DC |
153 | |
154 | Host Configuration | |
155 | ^^^^^^^^^^^^^^^^^^ | |
156 | ||
49f20f1b TL |
157 | In this case, the host cannot use the card. There are two methods to achieve |
158 | this: | |
6e4c46c4 | 159 | |
49f20f1b TL |
160 | * pass the device IDs to the options of the 'vfio-pci' modules by adding |
161 | + | |
162 | ---- | |
6e4c46c4 | 163 | options vfio-pci ids=1234:5678,4321:8765 |
6e4c46c4 | 164 | ---- |
49f20f1b TL |
165 | + |
166 | to a .conf file in */etc/modprobe.d/* where `1234:5678` and `4321:8765` are | |
167 | the vendor and device IDs obtained by: | |
168 | + | |
169 | ---- | |
170 | # lcpci -nn | |
6e4c46c4 DC |
171 | ---- |
172 | ||
49f20f1b TL |
173 | * blacklist the driver completely on the host, ensuring that it is free to bind |
174 | for passthrough, with | |
175 | + | |
176 | ---- | |
6e4c46c4 | 177 | blacklist DRIVERNAME |
49f20f1b TL |
178 | ---- |
179 | + | |
180 | in a .conf file in */etc/modprobe.d/*. | |
6e4c46c4 | 181 | |
49f20f1b TL |
182 | For both methods you need to |
183 | xref:qm_pci_passthrough_update_initramfs[update the `initramfs`] again and | |
184 | reboot after that. | |
6e4c46c4 | 185 | |
49f20f1b | 186 | [[qm_pci_passthrough_vm_config]] |
6e4c46c4 DC |
187 | VM Configuration |
188 | ^^^^^^^^^^^^^^^^ | |
49f20f1b TL |
189 | To pass through the device you need to set the *hostpciX* option in the VM |
190 | configuration, for example by executing: | |
6e4c46c4 DC |
191 | |
192 | ---- | |
49f20f1b | 193 | # qm set VMID -hostpci0 00:02.0 |
6e4c46c4 DC |
194 | ---- |
195 | ||
5ee3d3cd | 196 | If your device has multiple functions (e.g., ``00:02.0`' and ``00:02.1`' ), |
1c1241f2 | 197 | you can pass them through all together with the shortened syntax ``00:02`' |
6e4c46c4 DC |
198 | |
199 | There are some options to which may be necessary, depending on the device | |
49f20f1b TL |
200 | and guest OS: |
201 | ||
202 | * *x-vga=on|off* marks the PCI(e) device as the primary GPU of the VM. | |
203 | With this enabled the *vga* configuration option will be ignored. | |
6e4c46c4 | 204 | |
6e4c46c4 | 205 | * *pcie=on|off* tells {pve} to use a PCIe or PCI port. Some guests/device |
49f20f1b TL |
206 | combination require PCIe rather than PCI. PCIe is only available for 'q35' |
207 | machine types. | |
208 | ||
6e4c46c4 DC |
209 | * *rombar=on|off* makes the firmware ROM visible for the guest. Default is on. |
210 | Some PCI(e) devices need this disabled. | |
49f20f1b | 211 | |
6e4c46c4 | 212 | * *romfile=<path>*, is an optional path to a ROM file for the device to use. |
49f20f1b TL |
213 | This is a relative path under */usr/share/kvm/*. |
214 | ||
39d84f28 | 215 | .Example |
6e4c46c4 DC |
216 | |
217 | An example of PCIe passthrough with a GPU set to primary: | |
218 | ||
219 | ---- | |
49f20f1b | 220 | # qm set VMID -hostpci0 02:00,pcie=on,x-vga=on |
6e4c46c4 DC |
221 | ---- |
222 | ||
49f20f1b | 223 | |
6e4c46c4 DC |
224 | Other considerations |
225 | ^^^^^^^^^^^^^^^^^^^^ | |
226 | ||
227 | When passing through a GPU, the best compatibility is reached when using | |
49f20f1b TL |
228 | 'q35' as machine type, 'OVMF' ('EFI' for VMs) instead of SeaBIOS and PCIe |
229 | instead of PCI. Note that if you want to use 'OVMF' for GPU passthrough, the | |
230 | GPU needs to have an EFI capable ROM, otherwise use SeaBIOS instead. | |
6e4c46c4 DC |
231 | |
232 | SR-IOV | |
233 | ~~~~~~ | |
234 | ||
49f20f1b TL |
235 | Another variant for passing through PCI(e) devices, is to use the hardware |
236 | virtualization features of your devices, if available. | |
237 | ||
238 | 'SR-IOV' (**S**ingle-**R**oot **I**nput/**O**utput **V**irtualization) enables | |
239 | a single device to provide multiple 'VF' (**V**irtual **F**unctions) to the | |
240 | system. Each of those 'VF' can be used in a different VM, with full hardware | |
241 | features and also better performance and lower latency than software | |
242 | virtualized devices. | |
6e4c46c4 | 243 | |
49f20f1b TL |
244 | Currently, the most common use case for this are NICs (**N**etwork |
245 | **I**nterface **C**ard) with SR-IOV support, which can provide multiple VFs per | |
246 | physical port. This allows using features such as checksum offloading, etc. to | |
247 | be used inside a VM, reducing the (host) CPU overhead. | |
6e4c46c4 | 248 | |
6e4c46c4 DC |
249 | |
250 | Host Configuration | |
251 | ^^^^^^^^^^^^^^^^^^ | |
252 | ||
49f20f1b | 253 | Generally, there are two methods for enabling virtual functions on a device. |
6e4c46c4 | 254 | |
49f20f1b | 255 | * sometimes there is an option for the driver module e.g. for some |
6e4c46c4 | 256 | Intel drivers |
49f20f1b TL |
257 | + |
258 | ---- | |
6e4c46c4 | 259 | max_vfs=4 |
49f20f1b TL |
260 | ---- |
261 | + | |
262 | which could be put file with '.conf' ending under */etc/modprobe.d/*. | |
6e4c46c4 | 263 | (Do not forget to update your initramfs after that) |
49f20f1b | 264 | + |
6e4c46c4 DC |
265 | Please refer to your driver module documentation for the exact |
266 | parameters and options. | |
267 | ||
49f20f1b TL |
268 | * The second, more generic, approach is using the `sysfs`. |
269 | If a device and driver supports this you can change the number of VFs on | |
270 | the fly. For example, to setup 4 VFs on device 0000:01:00.0 execute: | |
271 | + | |
6e4c46c4 | 272 | ---- |
49f20f1b | 273 | # echo 4 > /sys/bus/pci/devices/0000:01:00.0/sriov_numvfs |
6e4c46c4 | 274 | ---- |
49f20f1b TL |
275 | + |
276 | To make this change persistent you can use the `sysfsutils` Debian package. | |
39d84f28 | 277 | After installation configure it via */etc/sysfs.conf* or a `FILE.conf' in |
49f20f1b | 278 | */etc/sysfs.d/*. |
6e4c46c4 DC |
279 | |
280 | VM Configuration | |
281 | ^^^^^^^^^^^^^^^^ | |
282 | ||
49f20f1b TL |
283 | After creating VFs, you should see them as separate PCI(e) devices when |
284 | outputting them with `lspci`. Get their ID and pass them through like a | |
285 | xref:qm_pci_passthrough_vm_config[normal PCI(e) device]. | |
6e4c46c4 DC |
286 | |
287 | Other considerations | |
288 | ^^^^^^^^^^^^^^^^^^^^ | |
289 | ||
290 | For this feature, platform support is especially important. It may be necessary | |
49f20f1b TL |
291 | to enable this feature in the BIOS/EFI first, or to use a specific PCI(e) port |
292 | for it to work. In doubt, consult the manual of the platform or contact its | |
293 | vendor. | |
050192c5 | 294 | |
d25f097c TL |
295 | Mediated Devices (vGPU, GVT-g) |
296 | ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ | |
050192c5 | 297 | |
d25f097c TL |
298 | Mediated devices are another method to use reuse features and performance from |
299 | physical hardware for virtualized hardware. These are found most common in | |
300 | virtualized GPU setups such as Intels GVT-g and Nvidias vGPUs used in their | |
301 | GRID technology. | |
302 | ||
303 | With this, a physical Card is able to create virtual cards, similar to SR-IOV. | |
304 | The difference is that mediated devices do not appear as PCI(e) devices in the | |
305 | host, and are such only suited for using in virtual machines. | |
050192c5 | 306 | |
050192c5 DC |
307 | |
308 | Host Configuration | |
309 | ^^^^^^^^^^^^^^^^^^ | |
310 | ||
d25f097c | 311 | In general your card's driver must support that feature, otherwise it will |
050192c5 DC |
312 | not work. So please refer to your vendor for compatbile drivers and how to |
313 | configure them. | |
314 | ||
315 | Intels drivers for GVT-g are integraded in the Kernel and should work | |
d25f097c TL |
316 | with the 5th, 6th and 7th generation Intel Core Processors, further E3 v4, E3 |
317 | v5 and E3 v6 Xeon Processors are supported. | |
050192c5 DC |
318 | |
319 | To enable it for Intel Graphcs, you have to make sure to load the module | |
320 | 'kvmgt' (for example via `/etc/modules`) and to enable it on the Kernel | |
d25f097c TL |
321 | commandline. For this you can edit `'/etc/default/grub'' and add the following |
322 | to the 'GRUB_CMDLINE_LINUX_DEFAULT' variable: | |
050192c5 DC |
323 | |
324 | ---- | |
325 | i915.enable_gvt=1 | |
326 | ---- | |
327 | ||
328 | After that remember to | |
329 | xref:qm_pci_passthrough_update_initramfs[update the `initramfs`], | |
330 | xref:qm_pci_passthrough_update_grub[update grub] and | |
331 | reboot your host. | |
332 | ||
333 | VM Configuration | |
334 | ^^^^^^^^^^^^^^^^ | |
335 | ||
d25f097c TL |
336 | To use a mediated device, simply specify the `mdev` property on a `hostpciX` |
337 | VM configuration option. | |
050192c5 | 338 | |
d25f097c TL |
339 | You can get the supported devices via the 'sysfs'. For example, to list the |
340 | supported types for the device '0000:00:02.0' you would simply execute: | |
050192c5 DC |
341 | |
342 | ---- | |
343 | # ls /sys/bus/pci/devices/0000:00:02.0/mdev_supported_types | |
344 | ---- | |
345 | ||
346 | Each entry is a directory which contains the following important files: | |
347 | ||
d25f097c TL |
348 | * 'available_instances' contains the amount of still available instances of |
349 | this type, each 'mdev' use in a VM reduces this. | |
050192c5 | 350 | * 'description' contains a short description about the capabilities of the type |
d25f097c TL |
351 | * 'create' is the endpoint to create such a device, {pve} does this |
352 | automatically for you, if a 'hostpciX' option with `mdev` is configured. | |
050192c5 | 353 | |
d25f097c | 354 | Example configuration with an `Intel GVT-g vGPU` (`Intel Skylake 6700k`): |
050192c5 DC |
355 | |
356 | ---- | |
357 | # qm set VMID -hostpci0 00:02.0,mdev=i915-GVTg_V5_4 | |
358 | ---- | |
359 | ||
360 | With this set, {pve} automatically creates such a device on VM start, and | |
361 | cleans it up again when the VM stops. | |
e582833b DC |
362 | |
363 | ifdef::wiki[] | |
364 | ||
365 | See Also | |
366 | ~~~~~~~~ | |
367 | ||
368 | * link:/wiki/Pci_passthrough[PCI Passthrough Examples] | |
369 | ||
370 | endif::wiki[] |