Table Of Contents
- API Methods
- acme.dns.authenticator
- alert
- alertclasses
- alertservice
- api_key
- app
- app.available
- app.available_space
- app.categories
- app.certificate_authority_choices
- app.certificate_choices
- app.config
- app.container_console_choices
- app.container_ids
- app.convert_to_custom
- app.create
- app.delete
- app.get_instance
- app.gpu_choices
- app.ip_choices
- app.latest
- app.outdated_docker_images
- app.pull_images
- app.query
- app.redeploy
- app.rollback
- app.rollback_versions
- app.similar
- app.start
- app.stop
- app.update
- app.upgrade
- app.upgrade_summary
- app.used_ports
- app.image
- app.ix_volume
- app.registry
- auth
- boot.environment
- catalog
- certificate
- certificateauthority
- cloud_backup
- cloudsync
- cloudsync.credentials
- config
- core
- cronjob
- device
- disk
- docker
- docker.network
- enclosure.label
- failover.reboot
- filesystem
- filesystem.acltemplate
- ftp
- group
- initshutdownscript
- ipmi
- iscsi.auth
- iscsi.extent
- iscsi.global
- iscsi.initiator
- iscsi.portal
- iscsi.target
- iscsi.targetextent
- k8s_to_docker
- keychaincredential
- keychaincredential.create
- keychaincredential.delete
- keychaincredential.generate_ssh_key_pair
- keychaincredential.get_instance
- keychaincredential.query
- keychaincredential.remote_ssh_host_key_scan
- keychaincredential.remote_ssh_semiautomatic_setup
- keychaincredential.setup_ssh_connection
- keychaincredential.update
- keychaincredential.used_by
- nfs
- pool
- pool.dataset
- pool.resilver
- pool.scrub
- pool.snapshottask
- pool.snapshottask.create
- pool.snapshottask.delete
- pool.snapshottask.delete_will_change_retention_for
- pool.snapshottask.get_instance
- pool.snapshottask.max_count
- pool.snapshottask.max_total_count
- pool.snapshottask.query
- pool.snapshottask.run
- pool.snapshottask.update
- pool.snapshottask.update_will_change_retention_for
- privilege
- reporting
- reporting.exporters
- sharing.nfs
- sharing.smb
- smb
- snmp
- staticroute
- system
- system.general
- system.ntpserver
- system.reboot
- system.security
- system.security.info
- tn_connect
- truenas
- user
- user.create
- user.delete
- user.get_instance
- user.get_next_uid
- user.get_user_obj
- user.has_local_administrator_set_up
- user.provisioning_uri
- user.query
- user.renew_2fa_secret
- user.set_password
- user.setup_local_administrator
- user.shell_choices
- user.twofactor_config
- user.unset_2fa_secret
- user.update
- user.verify_twofactor_token
- virt.device
- virt.global
- virt.instance
- virt.volume
- vm
- vm.bootloader_options
- vm.bootloader_ovmf_choices
- vm.clone
- vm.cpu_model_choices
- vm.create
- vm.delete
- vm.export_disk_image
- vm.flags
- vm.get_available_memory
- vm.get_console
- vm.get_display_devices
- vm.get_display_web_uri
- vm.get_instance
- vm.get_memory_usage
- vm.get_vm_memory_info
- vm.get_vmemory_in_use
- vm.guest_architecture_and_machine_choices
- vm.import_disk_image
- vm.log_file_download
- vm.log_file_path
- vm.maximum_supported_vcpus
- vm.port_wizard
- vm.poweroff
- vm.query
- vm.random_mac
- vm.resolution_choices
- vm.restart
- vm.resume
- vm.start
- vm.status
- vm.stop
- vm.supports_virtualization
- vm.suspend
- vm.update
- vm.virtualization_details
- vm.device
- vm.device.bind_choices
- vm.device.create
- vm.device.delete
- vm.device.disk_choices
- vm.device.get_instance
- vm.device.get_pci_ids_for_gpu_isolation
- vm.device.iommu_enabled
- vm.device.iotype_choices
- vm.device.nic_attach_choices
- vm.device.passthrough_device
- vm.device.passthrough_device_choices
- vm.device.pptdev_choices
- vm.device.query
- vm.device.update
- vm.device.usb_controller_choices
- vm.device.usb_passthrough_choices
- vm.device.usb_passthrough_device
- Jobs
- Query Methods
Previous topic
Next topic
vm.device.get_pci_ids_for_gpu_isolation¶
Get PCI IDs of devices which are required to be isolated for gpu_pci_id GPU isolation.
Basically when a GPU passthrough is desired for a VM, we need to isolate all the devices which are in the same IOMMU group as the GPU. This is required because if we don’t do this, the VM will not be able to start because the devices in the same IOMMU group as the GPU will be in use by the host and will not be available for the VM to use.
This endpoints retrieves all the PCI devices which are in the same IOMMU group as the GPU and returns their PCI IDs so UI can use those and create PCI devices for them and isolate them.
Type: object
No Additional Items
Type: array
No Additional Items
No Additional Items
Tuple Validation
Parameter 1: gpu_pci_id
gpu_pci_id
Type: stringMust be at least 1
characters long
Result
Type: array of stringNo Additional Items
Each item of this array must be:
Type: string
Must be at least 1
characters long
Required roles: VM_DEVICE_READ