The RUNNING_P2P state is designed to support multiple devices in the same
VM that are doing P2P transactions between themselves. When in RUNNING_P2P
the device must be able to accept incoming P2P transactions but should not
generate outgoing P2P transactions.
As an optional extension to the mandatory states it is defined as
in between STOP and RUNNING:
STOP -> RUNNING_P2P -> RUNNING -> RUNNING_P2P -> STOP
For drivers that are unable to support RUNNING_P2P the core code
silently merges RUNNING_P2P and RUNNING together. Unless driver support
is present, the new state cannot be used in SET_STATE.
Drivers that support this will be required to implement 4 FSM arcs
beyond the basic FSM. 2 of the basic FSM arcs become combination
transitions.
Compared to the v1 clarification, NDMA is redefined into FSM states and is
described in terms of the desired P2P quiescent behavior, noting that
halting all DMA is an acceptable implementation.
Link: https://lore.kernel.org/all/20220224142024.147653-11-yishaih@nvidia.com
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
Tested-by: Shameer Kolothum <shameerali.kolothum.thodi@huawei.com>
Reviewed-by: Kevin Tian <kevin.tian@intel.com>
Reviewed-by: Alex Williamson <alex.williamson@redhat.com>
Signed-off-by: Yishai Hadas <yishaih@nvidia.com>
Signed-off-by: Leon Romanovsky <leonro@nvidia.com>
Replace the existing region based migration protocol with an ioctl based
protocol. The two protocols have the same general semantic behaviors, but
the way the data is transported is changed.
This is the STOP_COPY portion of the new protocol, it defines the 5 states
for basic stop and copy migration and the protocol to move the migration
data in/out of the kernel.
Compared to the clarification of the v1 protocol Alex proposed:
https://lore.kernel.org/r/163909282574.728533.7460416142511440919.stgit@omen
This has a few deliberate functional differences:
- ERROR arcs allow the device function to remain unchanged.
- The protocol is not required to return to the original state on
transition failure. Instead userspace can execute an unwind back to
the original state, reset, or do something else without needing kernel
support. This simplifies the kernel design and should userspace choose
a policy like always reset, avoids doing useless work in the kernel
on error handling paths.
- PRE_COPY is made optional, userspace must discover it before using it.
This reflects the fact that the majority of drivers we are aware of
right now will not implement PRE_COPY.
- segmentation is not part of the data stream protocol, the receiver
does not have to reproduce the framing boundaries.
The hybrid FSM for the device_state is described as a Mealy machine by
documenting each of the arcs the driver is required to implement. Defining
the remaining set of old/new device_state transitions as 'combination
transitions' which are naturally defined as taking multiple FSM arcs along
the shortest path within the FSM's digraph allows a complete matrix of
transitions.
A new VFIO_DEVICE_FEATURE of VFIO_DEVICE_FEATURE_MIG_DEVICE_STATE is
defined to replace writing to the device_state field in the region. This
allows returning a brand new FD whenever the requested transition opens
a data transfer session.
The VFIO core code implements the new feature and provides a helper
function to the driver. Using the helper the driver only has to
implement 6 of the FSM arcs and the other combination transitions are
elaborated consistently from those arcs.
A new VFIO_DEVICE_FEATURE of VFIO_DEVICE_FEATURE_MIGRATION is defined to
report the capability for migration and indicate which set of states and
arcs are supported by the device. The FSM provides a lot of flexibility to
make backwards compatible extensions but the VFIO_DEVICE_FEATURE also
allows for future breaking extensions for scenarios that cannot support
even the basic STOP_COPY requirements.
The VFIO_DEVICE_FEATURE_MIG_DEVICE_STATE with the GET option (i.e.
VFIO_DEVICE_FEATURE_GET) can be used to read the current migration state
of the VFIO device.
Data transfer sessions are now carried over a file descriptor, instead of
the region. The FD functions for the lifetime of the data transfer
session. read() and write() transfer the data with normal Linux stream FD
semantics. This design allows future expansion to support poll(),
io_uring, and other performance optimizations.
The complicated mmap mode for data transfer is discarded as current qemu
doesn't take meaningful advantage of it, and the new qemu implementation
avoids substantially all the performance penalty of using a read() on the
region.
Link: https://lore.kernel.org/all/20220224142024.147653-10-yishaih@nvidia.com
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
Tested-by: Shameer Kolothum <shameerali.kolothum.thodi@huawei.com>
Reviewed-by: Kevin Tian <kevin.tian@intel.com>
Reviewed-by: Alex Williamson <alex.williamson@redhat.com>
Reviewed-by: Cornelia Huck <cohuck@redhat.com>
Signed-off-by: Yishai Hadas <yishaih@nvidia.com>
Signed-off-by: Leon Romanovsky <leonro@nvidia.com>
Revert the uAPI changes from the below commit with notice that these
regions and capabilities are no longer provided.
Fixes: b392a19891 ("vfio/pci: remove vfio_pci_nvlink2")
Reported-by: Greg Kurz <groug@kaod.org>
Signed-off-by: Alex Williamson <alex.williamson@redhat.com>
Reviewed-by: Cornelia Huck <cohuck@redhat.com>
Reviewed-by: Greg Kurz <groug@kaod.org>
Tested-by: Greg Kurz <groug@kaod.org>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Message-Id: <162014341432.3807030.11054087109120670135.stgit@omen>
This driver never had any open userspace (which for VFIO would include
VM kernel drivers) that use it, and thus should never have been added
by our normal userspace ABI rules.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Acked-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Message-Id: <20210326061311.1497642-2-hch@lst.de>
Signed-off-by: Alex Williamson <alex.williamson@redhat.com>
Define interfaces that allow the underlying memory object of an iova
range to be mapped to a new host virtual address in the host process:
- VFIO_DMA_UNMAP_FLAG_VADDR for VFIO_IOMMU_UNMAP_DMA
- VFIO_DMA_MAP_FLAG_VADDR flag for VFIO_IOMMU_MAP_DMA
- VFIO_UPDATE_VADDR extension for VFIO_CHECK_EXTENSION
Unmap vaddr invalidates the host virtual address in an iova range, and
blocks vfio translation of host virtual addresses. DMA to already-mapped
pages continues. Map vaddr updates the base VA and resumes translation.
See comments in uapi/linux/vfio.h for more details.
Signed-off-by: Steve Sistare <steven.sistare@oracle.com>
Reviewed-by: Cornelia Huck <cohuck@redhat.com>
Signed-off-by: Alex Williamson <alex.williamson@redhat.com>
For the UNMAP_DMA ioctl, delete all mappings if VFIO_DMA_UNMAP_FLAG_ALL
is set. Define the corresponding VFIO_UNMAP_ALL extension.
Signed-off-by: Steve Sistare <steven.sistare@oracle.com>
Reviewed-by: Cornelia Huck <cohuck@redhat.com>
Signed-off-by: Alex Williamson <alex.williamson@redhat.com>
The device is being unplugged, so pass the request to userspace to
ask for a graceful cleanup. This should free up the thread that
would otherwise loop waiting for the device to be fully released.
Signed-off-by: Eric Farman <farman@linux.ibm.com>
Reviewed-by: Cornelia Huck <cohuck@redhat.com>
Signed-off-by: Alex Williamson <alex.williamson@redhat.com>
Allow the VFIO_DEVICE_GET_INFO ioctl to include a capability chain.
Add a flag indicating capability chain support, and introduce the
definitions for the first set of capabilities which are specified to
s390 zPCI devices.
Signed-off-by: Matthew Rosato <mjrosato@linux.ibm.com>
Reviewed-by: Cornelia Huck <cohuck@redhat.com>
Signed-off-by: Alex Williamson <alex.williamson@redhat.com>
DPAA2 (Data Path Acceleration Architecture) consists in
mechanisms for processing Ethernet packets, queue management,
accelerators, etc.
The Management Complex (mc) is a hardware entity that manages the DPAA2
hardware resources. It provides an object-based abstraction for software
drivers to use the DPAA2 hardware. The MC mediates operations such as
create, discover, destroy of DPAA2 objects.
The MC provides memory-mapped I/O command interfaces (MC portals) which
DPAA2 software drivers use to operate on DPAA2 objects.
A DPRC is a container object that holds other types of DPAA2 objects.
Each object in the DPRC is a Linux device and bound to a driver.
The MC-bus driver is a platform driver (different from PCI or platform
bus). The DPRC driver does runtime management of a bus instance. It
performs the initial scan of the DPRC and handles changes in the DPRC
configuration (adding/removing objects).
All objects inside a container share the same hardware isolation
context, meaning that only an entire DPRC can be assigned to
a virtual machine.
When a container is assigned to a virtual machine, all the objects
within that container are assigned to that virtual machine.
The DPRC container assigned to the virtual machine is not allowed
to change contents (add/remove objects) by the guest. The restriction
is set by the host and enforced by the mc hardware.
The DPAA2 objects can be directly assigned to the guest. However
the MC portals (the memory mapped command interface to the MC) need
to be emulated because there are commands that configure the
interrupts and the isolation IDs which are virtual in the guest.
Example:
echo vfio-fsl-mc > /sys/bus/fsl-mc/devices/dprc.2/driver_override
echo dprc.2 > /sys/bus/fsl-mc/drivers/vfio-fsl-mc/bind
The dprc.2 is bound to the VFIO driver and all the objects within
dprc.2 are going to be bound to the VFIO driver.
This patch adds the infrastructure for VFIO support for fsl-mc
devices. Subsequent patches will add support for binding and secure
assigning these devices using VFIO.
More details about the DPAA2 objects can be found here:
Documentation/networking/device_drivers/freescale/dpaa2/overview.rst
Signed-off-by: Bharat Bhushan <Bharat.Bhushan@nxp.com>
Signed-off-by: Diana Craciun <diana.craciun@oss.nxp.com>
Reviewed-by: Eric Auger <eric.auger@redhat.com>
Signed-off-by: Alex Williamson <alex.williamson@redhat.com>
Commit 492855939b ("vfio/type1: Limit DMA mappings per container")
added the ability to limit the number of memory backed DMA mappings.
However on s390x, when lazy mapping is in use, we use a very large
number of concurrent mappings. Let's provide the current allowable
number of DMA mappings to userspace via the IOMMU info chain so that
userspace can take appropriate mitigation.
Signed-off-by: Matthew Rosato <mjrosato@linux.ibm.com>
Reviewed-by: Cornelia Huck <cohuck@redhat.com>
Signed-off-by: Alex Williamson <alex.williamson@redhat.com>
ID 1 is already used by the IOVA range capability, use ID 2.
Reported-by: Liu Yi L <yi.l.liu@intel.com>
Fixes: ad721705d0 ("vfio iommu: Add migration capability to report supported features")
Reviewed-by: Kirti Wankhede <kwankhede@nvidia.com>
Signed-off-by: Alex Williamson <alex.williamson@redhat.com>
Pull s390 updates from Vasily Gorbik:
- Add support for multi-function devices in pci code.
- Enable PF-VF linking for architectures using the pdev->no_vf_scan
flag (currently just s390).
- Add reipl from NVMe support.
- Get rid of critical section cleanup in entry.S.
- Refactor PNSO CHSC (perform network subchannel operation) in cio and
qeth.
- QDIO interrupts and error handling fixes and improvements, more
refactoring changes.
- Align ioremap() with generic code.
- Accept requests without the prefetch bit set in vfio-ccw.
- Enable path handling via two new regions in vfio-ccw.
- Other small fixes and improvements all over the code.
* tag 's390-5.8-1' of git://git.kernel.org/pub/scm/linux/kernel/git/s390/linux: (52 commits)
vfio-ccw: make vfio_ccw_regops variables declarations static
vfio-ccw: Add trace for CRW event
vfio-ccw: Wire up the CRW irq and CRW region
vfio-ccw: Introduce a new CRW region
vfio-ccw: Refactor IRQ handlers
vfio-ccw: Introduce a new schib region
vfio-ccw: Refactor the unregister of the async regions
vfio-ccw: Register a chp_event callback for vfio-ccw
vfio-ccw: Introduce new helper functions to free/destroy regions
vfio-ccw: document possible errors
vfio-ccw: Enable transparent CCW IPL from DASD
s390/pci: Log new handle in clp_disable_fh()
s390/cio, s390/qeth: cleanup PNSO CHSC
s390/qdio: remove q->first_to_kick
s390/qdio: fix up qdio_start_irq() kerneldoc
s390: remove critical section cleanup from entry.S
s390: add machine check SIGP
s390/pci: ioremap() align with generic code
s390/ap: introduce new ap function ap_get_qdev()
Documentation/s390: Update / remove developerWorks web links
...
This region provides a mechanism to pass a Channel Report Word
that affect vfio-ccw devices, and needs to be passed to the guest
for its awareness and/or processing.
The base driver (see crw_collect_info()) provides space for two
CRWs, as a subchannel event may have two CRWs chained together
(one for the ssid, one for the subchannel). As vfio-ccw will
deal with everything at the subchannel level, provide space
for a single CRW to be transferred in one shot.
Signed-off-by: Farhan Ali <alifm@linux.ibm.com>
Signed-off-by: Eric Farman <farman@linux.ibm.com>
Reviewed-by: Cornelia Huck <cohuck@redhat.com>
Message-Id: <20200505122745.53208-7-farman@linux.ibm.com>
[CH: added padding to ccw_crw_region]
Signed-off-by: Cornelia Huck <cohuck@redhat.com>
The schib region can be used by userspace to get the subchannel-
information block (SCHIB) for the passthrough subchannel.
This can be useful to get information such as channel path
information via the SCHIB.PMCW fields.
Signed-off-by: Farhan Ali <alifm@linux.ibm.com>
Signed-off-by: Eric Farman <farman@linux.ibm.com>
Reviewed-by: Cornelia Huck <cohuck@redhat.com>
Message-Id: <20200505122745.53208-5-farman@linux.ibm.com>
Signed-off-by: Cornelia Huck <cohuck@redhat.com>
Added migration capability in IOMMU info chain.
User application should check IOMMU info chain for migration capability
to use dirty page tracking feature provided by kernel module.
User application must check page sizes supported and maximum dirty
bitmap size returned by this capability structure for ioctls used to get
dirty bitmap.
Signed-off-by: Kirti Wankhede <kwankhede@nvidia.com>
Reviewed-by: Cornelia Huck <cohuck@redhat.com>
Reviewed-by: Yan Zhao <yan.y.zhao@intel.com>
Signed-off-by: Alex Williamson <alex.williamson@redhat.com>
DMA mapped pages, including those pinned by mdev vendor drivers, might
get unpinned and unmapped while migration is active and device is still
running. For example, in pre-copy phase while guest driver could access
those pages, host device or vendor driver can dirty these mapped pages.
Such pages should be marked dirty so as to maintain memory consistency
for a user making use of dirty page tracking.
To get bitmap during unmap, user should allocate memory for bitmap, set
it all zeros, set size of allocated memory, set page size to be
considered for bitmap and set flag VFIO_DMA_UNMAP_FLAG_GET_DIRTY_BITMAP.
Signed-off-by: Kirti Wankhede <kwankhede@nvidia.com>
Reviewed-by: Neo Jia <cjia@nvidia.com>
Reviewed-by: Cornelia Huck <cohuck@redhat.com>
Reviewed-by: Yan Zhao <yan.y.zhao@intel.com>
Signed-off-by: Alex Williamson <alex.williamson@redhat.com>
IOMMU container maintains a list of all pages pinned by vfio_pin_pages API.
All pages pinned by vendor driver through this API should be considered as
dirty during migration. When container consists of IOMMU capable device and
all pages are pinned and mapped, then all pages are marked dirty.
Added support to start/stop dirtied pages tracking and to get bitmap of all
dirtied pages for requested IO virtual address range.
Signed-off-by: Kirti Wankhede <kwankhede@nvidia.com>
Reviewed-by: Neo Jia <cjia@nvidia.com>
Reviewed-by: Cornelia Huck <cohuck@redhat.com>
Reviewed-by: Yan Zhao <yan.y.zhao@intel.com>
Signed-off-by: Alex Williamson <alex.williamson@redhat.com>
- Defined MIGRATION region type and sub-type.
- Defined vfio_device_migration_info structure which will be placed at the
0th offset of migration region to get/set VFIO device related
information. Defined members of structure and usage on read/write access.
- Defined device states and state transition details.
- Defined sequence to be followed while saving and resuming VFIO device.
Signed-off-by: Kirti Wankhede <kwankhede@nvidia.com>
Reviewed-by: Neo Jia <cjia@nvidia.com>
Reviewed-by: Cornelia Huck <cohuck@redhat.com>
Reviewed-by: Yan Zhao <yan.y.zhao@intel.com>
Signed-off-by: Alex Williamson <alex.williamson@redhat.com>
The VFIO_DEVICE_FEATURE ioctl is meant to be a general purpose, device
agnostic ioctl for setting, retrieving, and probing device features.
This implementation provides a 16-bit field for specifying a feature
index, where the data porition of the ioctl is determined by the
semantics for the given feature. Additional flag bits indicate the
direction and nature of the operation; SET indicates user data is
provided into the device feature, GET indicates the device feature is
written out into user data. The PROBE flag augments determining
whether the given feature is supported, and if provided, whether the
given operation on the feature is supported.
The first user of this ioctl is for setting the vfio-pci VF token,
where the user provides a shared secret key (UUID) on a SR-IOV PF
device, which users must provide when opening associated VF devices.
Reviewed-by: Cornelia Huck <cohuck@redhat.com>
Reviewed-by: Kevin Tian <kevin.tian@intel.com>
Signed-off-by: Alex Williamson <alex.williamson@redhat.com>
