devel/migration/vfio.rst

8 devices is done in QEMU.
15 helps to reduce the total downtime of the VM. VFIO devices opt-in to pre-copy
16 support by reporting the VFIO_MIGRATION_PRE_COPY flag in the
22 and recommends that the initial bytes are sent and loaded in the destination
28 While in the P2P quiescent state, P2P DMA transactions cannot be initiated by
42 A detailed description of the UAPI for VFIO device migration can be found in
43 the comment for the ``vfio_device_mig_state`` structure in the header file
50 * A ``load_setup`` function that sets the VFIO device on the destination in
62   active only when the VFIO device is in pre-copy states.
70 * A ``switchover_start`` function that in the multifd mode starts a thread that
71   reassembles the multifd received data and loads it in-order into the device.
75   in the non-multifd mode.
78 * A ``save_live_complete_precopy`` function that sets the VFIO device in
83 * A ``save_live_complete_precopy_thread`` function that in the multifd mode
97   It's used only in the multifd mode.
109 example, the VFIO device state is transitioned back to _RUNNING in case a
127 More info about the uAPI can be found in the comments of the
129 ``vfio_device_feature_dma_logging_report`` structures in the header file
136 APIs in which case only those pinned pages are perpetually marked dirty.
142 phase. So, a page marked as dirty will be copied to the destination in both
143 phases. Copying dirty pages in pre-copy phase helps QEMU to predict if it can
145 dirty pages continuously, then it understands that even in stop-and-copy phase,
150 off, all dirty pages will be copied to the destination in stop-and-copy phase
156 With vIOMMU, an IO virtual address range can get unmapped while in pre-copy
157 phase of migration. In that case, the unmap ioctl returns any dirty pages in
171 The values in the parentheses represent the VM state, the migration state, and
193               The VFIO device is first put in P2P quiescent state
196                 Then the VFIO device is put in _STOP_COPY state
199               For the VFIO device: in the non-multifd mode iterate in
202 	          In the multifd mode this iteration is done in
226               The VFIO device is first put in P2P quiescent state
243 saving its config space is also parallelized (run in a separate thread) in
249 attempted in configurations that otherwise support it.