pci, vmm: Defer mapping VFIO MMIO regions on restore

When restoring a VM, the restore codepath will take care of mapping the
MMIO regions based on the information from the snapshot, rather than
having the mapping being performed during device creation.

When the device is created, information such as which BARs contain the
MSI-X tables are missing, preventing to perform the mapping of the MMIO
regions.

Signed-off-by: Sebastien Boeuf <sebastien.boeuf@intel.com>
This commit is contained in:
Sebastien Boeuf
2022-06-08 16:41:24 +02:00
parent 7df7061610
commit 81ba70a497
3 changed files with 57 additions and 40 deletions

View File

@@ -16,8 +16,7 @@ use crate::config::{
use crate::device_tree::{DeviceNode, DeviceTree};
use crate::interrupt::LegacyUserspaceInterruptManager;
use crate::interrupt::MsiInterruptManager;
use crate::memory_manager::MEMORY_MANAGER_ACPI_SIZE;
use crate::memory_manager::{Error as MemoryManagerError, MemoryManager};
use crate::memory_manager::{Error as MemoryManagerError, MemoryManager, MEMORY_MANAGER_ACPI_SIZE};
use crate::pci_segment::PciSegment;
use crate::seccomp_filters::{get_seccomp_filter, Thread};
use crate::serial_manager::{Error as SerialManagerError, SerialManager};
@@ -3074,6 +3073,8 @@ impl DeviceManager {
None
};
let memory_manager = self.memory_manager.clone();
let vfio_pci_device = VfioPciDevice::new(
vfio_name.clone(),
&self.address_manager.vm,
@@ -3084,6 +3085,7 @@ impl DeviceManager {
device_cfg.iommu,
pci_device_bdf,
self.restoring,
Arc::new(move || memory_manager.lock().unwrap().allocate_memory_slot()),
)
.map_err(DeviceManagerError::VfioPciCreate)?;
@@ -3097,13 +3099,15 @@ impl DeviceManager {
resources,
)?;
vfio_pci_device
.lock()
.unwrap()
.map_mmio_regions(&self.address_manager.vm, || {
self.memory_manager.lock().unwrap().allocate_memory_slot()
})
.map_err(DeviceManagerError::VfioMapRegion)?;
// When restoring a VM, the restore codepath will take care of mapping
// the MMIO regions based on the information from the snapshot.
if !self.restoring {
vfio_pci_device
.lock()
.unwrap()
.map_mmio_regions()
.map_err(DeviceManagerError::VfioMapRegion)?;
}
let mut node = device_node!(vfio_name, vfio_pci_device);
@@ -3230,6 +3234,8 @@ impl DeviceManager {
.map_err(DeviceManagerError::VfioUserCreateClient)?,
));
let memory_manager = self.memory_manager.clone();
let mut vfio_user_pci_device = VfioUserPciDevice::new(
vfio_user_name.clone(),
&self.address_manager.vm,
@@ -3238,6 +3244,7 @@ impl DeviceManager {
legacy_interrupt_group,
pci_device_bdf,
self.restoring,
Arc::new(move || memory_manager.lock().unwrap().allocate_memory_slot()),
)
.map_err(DeviceManagerError::VfioUserCreate)?;
@@ -3272,15 +3279,17 @@ impl DeviceManager {
resources,
)?;
// Note it is required to call 'add_pci_device()' in advance to have the list of
// mmio regions provisioned correctly
vfio_user_pci_device
.lock()
.unwrap()
.map_mmio_regions(&self.address_manager.vm, || {
self.memory_manager.lock().unwrap().allocate_memory_slot()
})
.map_err(DeviceManagerError::VfioUserMapRegion)?;
// When restoring a VM, the restore codepath will take care of mapping
// the MMIO regions based on the information from the snapshot.
if !self.restoring {
// Note it is required to call 'add_pci_device()' in advance to have the list of
// mmio regions provisioned correctly
vfio_user_pci_device
.lock()
.unwrap()
.map_mmio_regions()
.map_err(DeviceManagerError::VfioUserMapRegion)?;
}
let mut node = device_node!(vfio_user_name, vfio_user_pci_device);