1===================== 2VFIO device migration 3===================== 4 5Migration of virtual machine involves saving the state for each device that 6the guest is running on source host and restoring this saved state on the 7destination host. This document details how saving and restoring of VFIO 8devices is done in QEMU. 9 10Migration of VFIO devices consists of two phases: the optional pre-copy phase, 11and the stop-and-copy phase. The pre-copy phase is iterative and allows to 12accommodate VFIO devices that have a large amount of data that needs to be 13transferred. The iterative pre-copy phase of migration allows for the guest to 14continue whilst the VFIO device state is transferred to the destination, this 15helps to reduce the total downtime of the VM. VFIO devices opt-in to pre-copy 16support by reporting the VFIO_MIGRATION_PRE_COPY flag in the 17VFIO_DEVICE_FEATURE_MIGRATION ioctl. 18 19When pre-copy is supported, it's possible to further reduce downtime by 20enabling "switchover-ack" migration capability. 21VFIO migration uAPI defines "initial bytes" as part of its pre-copy data stream 22and recommends that the initial bytes are sent and loaded in the destination 23before stopping the source VM. Enabling this migration capability will 24guarantee that and thus, can potentially reduce downtime even further. 25 26To support migration of multiple devices that might do P2P transactions between 27themselves, VFIO migration uAPI defines an intermediate P2P quiescent state. 28While in the P2P quiescent state, P2P DMA transactions cannot be initiated by 29the device, but the device can respond to incoming ones. Additionally, all 30outstanding P2P transactions are guaranteed to have been completed by the time 31the device enters this state. 32 33All the devices that support P2P migration are first transitioned to the P2P 34quiescent state and only then are they stopped or started. This makes migration 35safe P2P-wise, since starting and stopping the devices is not done atomically 36for all the devices together. 37 38Thus, multiple VFIO devices migration is allowed only if all the devices 39support P2P migration. Single VFIO device migration is allowed regardless of 40P2P migration support. 41 42A detailed description of the UAPI for VFIO device migration can be found in 43the comment for the ``vfio_device_mig_state`` structure in the header file 44linux-headers/linux/vfio.h. 45 46VFIO implements the device hooks for the iterative approach as follows: 47 48* A ``save_setup`` function that sets up migration on the source. 49 50* A ``load_setup`` function that sets the VFIO device on the destination in 51 _RESUMING state. 52 53* A ``state_pending_estimate`` function that reports an estimate of the 54 remaining pre-copy data that the vendor driver has yet to save for the VFIO 55 device. 56 57* A ``state_pending_exact`` function that reads pending_bytes from the vendor 58 driver, which indicates the amount of data that the vendor driver has yet to 59 save for the VFIO device. 60 61* An ``is_active_iterate`` function that indicates ``save_live_iterate`` is 62 active only when the VFIO device is in pre-copy states. 63 64* A ``save_live_iterate`` function that reads the VFIO device's data from the 65 vendor driver during iterative pre-copy phase. 66 67* A ``switchover_ack_needed`` function that checks if the VFIO device uses 68 "switchover-ack" migration capability when this capability is enabled. 69 70* A ``save_state`` function to save the device config space if it is present. 71 72* A ``save_live_complete_precopy`` function that sets the VFIO device in 73 _STOP_COPY state and iteratively copies the data for the VFIO device until 74 the vendor driver indicates that no data remains. 75 76* A ``load_state`` function that loads the config section and the data 77 sections that are generated by the save functions above. 78 79* A ``load_state_buffer`` function that loads the device state and the device 80 config that arrived via multifd channels. 81 It's used only in the multifd mode. 82 83* ``cleanup`` functions for both save and load that perform any migration 84 related cleanup. 85 86 87The VFIO migration code uses a VM state change handler to change the VFIO 88device state when the VM state changes from running to not-running, and 89vice versa. 90 91Similarly, a migration state change handler is used to trigger a transition of 92the VFIO device state when certain changes of the migration state occur. For 93example, the VFIO device state is transitioned back to _RUNNING in case a 94migration failed or was canceled. 95 96System memory dirty pages tracking 97---------------------------------- 98 99A ``log_global_start`` and ``log_global_stop`` memory listener callback informs 100the VFIO dirty tracking module to start and stop dirty page tracking. A 101``log_sync`` memory listener callback queries the dirty page bitmap from the 102dirty tracking module and marks system memory pages which were DMA-ed by the 103VFIO device as dirty. The dirty page bitmap is queried per container. 104 105Currently there are two ways dirty page tracking can be done: 106(1) Device dirty tracking: 107In this method the device is responsible to log and report its DMAs. This 108method can be used only if the device is capable of tracking its DMAs. 109Discovering device capability, starting and stopping dirty tracking, and 110syncing the dirty bitmaps from the device are done using the DMA logging uAPI. 111More info about the uAPI can be found in the comments of the 112``vfio_device_feature_dma_logging_control`` and 113``vfio_device_feature_dma_logging_report`` structures in the header file 114linux-headers/linux/vfio.h. 115 116(2) VFIO IOMMU module: 117In this method dirty tracking is done by IOMMU. However, there is currently no 118IOMMU support for dirty page tracking. For this reason, all pages are 119perpetually marked dirty, unless the device driver pins pages through external 120APIs in which case only those pinned pages are perpetually marked dirty. 121 122If the above two methods are not supported, all pages are perpetually marked 123dirty by QEMU. 124 125By default, dirty pages are tracked during pre-copy as well as stop-and-copy 126phase. So, a page marked as dirty will be copied to the destination in both 127phases. Copying dirty pages in pre-copy phase helps QEMU to predict if it can 128achieve its downtime tolerances. If QEMU during pre-copy phase keeps finding 129dirty pages continuously, then it understands that even in stop-and-copy phase, 130it is likely to find dirty pages and can predict the downtime accordingly. 131 132QEMU also provides a per device opt-out option ``pre-copy-dirty-page-tracking`` 133which disables querying the dirty bitmap during pre-copy phase. If it is set to 134off, all dirty pages will be copied to the destination in stop-and-copy phase 135only. 136 137System memory dirty pages tracking when vIOMMU is enabled 138--------------------------------------------------------- 139 140With vIOMMU, an IO virtual address range can get unmapped while in pre-copy 141phase of migration. In that case, the unmap ioctl returns any dirty pages in 142that range and QEMU reports corresponding guest physical pages dirty. During 143stop-and-copy phase, an IOMMU notifier is used to get a callback for mapped 144pages and then dirty pages bitmap is fetched from VFIO IOMMU modules for those 145mapped ranges. If device dirty tracking is enabled with vIOMMU, live migration 146will be blocked. 147 148Flow of state changes during Live migration 149=========================================== 150 151Below is the state change flow during live migration for a VFIO device that 152supports both precopy and P2P migration. The flow for devices that don't 153support it is similar, except that the relevant states for precopy and P2P are 154skipped. 155The values in the parentheses represent the VM state, the migration state, and 156the VFIO device state, respectively. 157 158Live migration save path 159------------------------ 160 161:: 162 163 QEMU normal running state 164 (RUNNING, _NONE, _RUNNING) 165 | 166 migrate_init spawns migration_thread 167 Migration thread then calls each device's .save_setup() 168 (RUNNING, _SETUP, _PRE_COPY) 169 | 170 (RUNNING, _ACTIVE, _PRE_COPY) 171 If device is active, get pending_bytes by .state_pending_{estimate,exact}() 172 If total pending_bytes >= threshold_size, call .save_live_iterate() 173 Data of VFIO device for pre-copy phase is copied 174 Iterate till total pending bytes converge and are less than threshold 175 | 176 On migration completion, the vCPUs and the VFIO device are stopped 177 The VFIO device is first put in P2P quiescent state 178 (FINISH_MIGRATE, _ACTIVE, _PRE_COPY_P2P) 179 | 180 Then the VFIO device is put in _STOP_COPY state 181 (FINISH_MIGRATE, _ACTIVE, _STOP_COPY) 182 .save_live_complete_precopy() is called for each active device 183 For the VFIO device, iterate in .save_live_complete_precopy() until 184 pending data is 0 185 | 186 (POSTMIGRATE, _COMPLETED, _STOP_COPY) 187 Migraton thread schedules cleanup bottom half and exits 188 | 189 .save_cleanup() is called 190 (POSTMIGRATE, _COMPLETED, _STOP) 191 192Live migration resume path 193-------------------------- 194 195:: 196 197 Incoming migration calls .load_setup() for each device 198 (RESTORE_VM, _ACTIVE, _STOP) 199 | 200 For each device, .load_state() is called for that device section data 201 transmitted via the main migration channel. 202 For data transmitted via multifd channels .load_state_buffer() is called 203 instead. 204 (RESTORE_VM, _ACTIVE, _RESUMING) 205 | 206 At the end, .load_cleanup() is called for each device and vCPUs are started 207 The VFIO device is first put in P2P quiescent state 208 (RUNNING, _ACTIVE, _RUNNING_P2P) 209 | 210 (RUNNING, _NONE, _RUNNING) 211 212Postcopy 213======== 214 215Postcopy migration is currently not supported for VFIO devices. 216