hisi_acc_vfio_pci: Update migration data pointer correctly on saving/resume

[ Upstream commit be12ad45e15b5ee0e2526a50266ba1d295d26a88 ]

When the optional PRE_COPY support was added to speed up the

hisi_acc_vfio_pci: Update migration data pointer correctly on saving/resume

[ Upstream commit be12ad45e15b5ee0e2526a50266ba1d295d26a88 ]

When the optional PRE_COPY support was added to speed up the device
compatibility check, it failed to update the saving/resuming data
pointers based on the fd offset. This results in migration data
corruption and when the device gets started on the destination the
following error is reported in some cases,

[ 478.907684] arm-smmu-v3 arm-smmu-v3.2.auto: event 0x10 received:
[ 478.913691] arm-smmu-v3 arm-smmu-v3.2.auto: 0x0000310200000010
[ 478.919603] arm-smmu-v3 arm-smmu-v3.2.auto: 0x000002088000007f
[ 478.925515] arm-smmu-v3 arm-smmu-v3.2.auto: 0x0000000000000000
[ 478.931425] arm-smmu-v3 arm-smmu-v3.2.auto: 0x0000000000000000
[ 478.947552] hisi_zip 0000:31:00.0: qm_axi_rresp [error status=0x1] found
[ 478.955930] hisi_zip 0000:31:00.0: qm_db_timeout [error status=0x400] found
[ 478.955944] hisi_zip 0000:31:00.0: qm sq doorbell timeout in function 2

Fixes: d9a871e4a143 ("hisi_acc_vfio_pci: Introduce support for PRE_COPY state transitions")
Signed-off-by: Shameer Kolothum <shameerali.kolothum.thodi@huawei.com>
Reviewed-by: Jason Gunthorpe <jgg@nvidia.com>
Link: https://lore.kernel.org/r/20231120091406.780-1-shameerali.kolothum.thodi@huawei.com
Signed-off-by: Alex Williamson <alex.williamson@redhat.com>
Signed-off-by: Sasha Levin <sashal@kernel.org>

show more ...

vfio-iommufd: Add detach_ioas support for physical VFIO devices

This prepares for adding DETACH ioctl for physical VFIO devices.

Reviewed-by: Kevin Tian <kevin.tian@intel.com>
Reviewed-by: Jason Gu

vfio-iommufd: Add detach_ioas support for physical VFIO devices

This prepares for adding DETACH ioctl for physical VFIO devices.

Reviewed-by: Kevin Tian <kevin.tian@intel.com>
Reviewed-by: Jason Gunthorpe <jgg@nvidia.com>
Tested-by: Terrence Xu <terrence.xu@intel.com>
Tested-by: Nicolin Chen <nicolinc@nvidia.com>
Tested-by: Matthew Rosato <mjrosato@linux.ibm.com>
Tested-by: Yanting Jiang <yanting.jiang@intel.com>
Tested-by: Shameer Kolothum <shameerali.kolothum.thodi@huawei.com>
Tested-by: Zhenzhong Duan <zhenzhong.duan@intel.com>
Signed-off-by: Yi Liu <yi.l.liu@intel.com>
Link: https://lore.kernel.org/r/20230718135551.6592-14-yi.l.liu@intel.com
Signed-off-by: Alex Williamson <alex.williamson@redhat.com>

show more ...

vfio/hisi: Use GFP_KERNEL_ACCOUNT for userspace persistent allocations

Use GFP_KERNEL_ACCOUNT for userspace persistent allocations.

The GFP_KERNEL_ACCOUNT option lets the memory allocator know that

vfio/hisi: Use GFP_KERNEL_ACCOUNT for userspace persistent allocations

Use GFP_KERNEL_ACCOUNT for userspace persistent allocations.

The GFP_KERNEL_ACCOUNT option lets the memory allocator know that this
is untrusted allocation triggered from userspace and should be a subject
of kmem accounting, and as such it is controlled by the cgroup
mechanism.

Signed-off-by: Yishai Hadas <yishaih@nvidia.com>
Reviewed-by: Jason Gunthorpe <jgg@nvidia.com>
Link: https://lore.kernel.org/r/20230108154427.32609-5-yishaih@nvidia.com
Signed-off-by: Alex Williamson <alex.williamson@redhat.com>

show more ...

hisi_acc_vfio_pci: Enable PRE_COPY flag

Now that we have everything to support the PRE_COPY state,
enable it.

Signed-off-by: Shameer Kolothum <shameerali.kolothum.thodi@huawei.com>
Link: https://lo

hisi_acc_vfio_pci: Enable PRE_COPY flag

Now that we have everything to support the PRE_COPY state,
enable it.

Signed-off-by: Shameer Kolothum <shameerali.kolothum.thodi@huawei.com>
Link: https://lore.kernel.org/r/20221123113236.896-5-shameerali.kolothum.thodi@huawei.com
Signed-off-by: Alex Williamson <alex.williamson@redhat.com>

show more ...

hisi_acc_vfio_pci: Move the dev compatibility tests for early check

Instead of waiting till data transfer is complete to perform dev
compatibility, do it as soon as we have enough data to perform th

hisi_acc_vfio_pci: Move the dev compatibility tests for early check

Instead of waiting till data transfer is complete to perform dev
compatibility, do it as soon as we have enough data to perform the
check. This will be useful when we enable the support for PRE_COPY.

Signed-off-by: Shameer Kolothum <shameerali.kolothum.thodi@huawei.com>
Link: https://lore.kernel.org/r/20221123113236.896-4-shameerali.kolothum.thodi@huawei.com
Signed-off-by: Alex Williamson <alex.williamson@redhat.com>

show more ...

hisi_acc_vfio_pci: Introduce support for PRE_COPY state transitions

The saving_migf is open in PRE_COPY state if it is supported and reads
initial device match data. hisi_acc_vf_stop_copy() is refac

hisi_acc_vfio_pci: Introduce support for PRE_COPY state transitions

The saving_migf is open in PRE_COPY state if it is supported and reads
initial device match data. hisi_acc_vf_stop_copy() is refactored to
make use of common code.

Signed-off-by: Shameer Kolothum <shameerali.kolothum.thodi@huawei.com>
Link: https://lore.kernel.org/r/20221123113236.896-3-shameerali.kolothum.thodi@huawei.com
Signed-off-by: Alex Williamson <alex.williamson@redhat.com>

show more ...

hisi_acc_vfio_pci: Add support for precopy IOCTL

PRECOPY IOCTL in the case of HiSiIicon ACC driver can be used to
perform the device compatibility check earlier during migration.

Signed-off-by: Sha

hisi_acc_vfio_pci: Add support for precopy IOCTL

PRECOPY IOCTL in the case of HiSiIicon ACC driver can be used to
perform the device compatibility check earlier during migration.

Signed-off-by: Shameer Kolothum <shameerali.kolothum.thodi@huawei.com>
Link: https://lore.kernel.org/r/20221123113236.896-2-shameerali.kolothum.thodi@huawei.com
Signed-off-by: Alex Williamson <alex.williamson@redhat.com>

show more ...

vfio-iommufd: Support iommufd for physical VFIO devices

This creates the iommufd_device for the physical VFIO drivers. These are
all the drivers that are calling vfio_register_group_dev() and expect

vfio-iommufd: Support iommufd for physical VFIO devices

This creates the iommufd_device for the physical VFIO drivers. These are
all the drivers that are calling vfio_register_group_dev() and expect the
type1 code to setup a real iommu_domain against their parent struct
device.

The design gives the driver a choice in how it gets connected to iommufd
by providing bind_iommufd/unbind_iommufd/attach_ioas callbacks to
implement as required. The core code provides three default callbacks for
physical mode using a real iommu_domain. This is suitable for drivers
using vfio_register_group_dev()

Link: https://lore.kernel.org/r/6-v4-42cd2eb0e3eb+335a-vfio_iommufd_jgg@nvidia.com
Reviewed-by: Kevin Tian <kevin.tian@intel.com>
Reviewed-by: Alex Williamson <alex.williamson@redhat.com>
Tested-by: Alex Williamson <alex.williamson@redhat.com>
Tested-by: Nicolin Chen <nicolinc@nvidia.com>
Tested-by: Yi Liu <yi.l.liu@intel.com>
Tested-by: Lixiao Yang <lixiao.yang@intel.com>
Tested-by: Matthew Rosato <mjrosato@linux.ibm.com>
Tested-by: Yu He <yu.he@intel.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>

show more ...

vfio: Add an option to get migration data size

Add an option to get migration data size by introducing a new migration
feature named VFIO_DEVICE_FEATURE_MIG_DATA_SIZE.

Upon VFIO_DEVICE_FEATURE_GET

vfio: Add an option to get migration data size

Add an option to get migration data size by introducing a new migration
feature named VFIO_DEVICE_FEATURE_MIG_DATA_SIZE.

Upon VFIO_DEVICE_FEATURE_GET the estimated data length that will be
required to complete STOP_COPY is returned.

This option may better enable user space to consider before moving to
STOP_COPY whether it can meet the downtime SLA based on the returned
data.

The patch also includes the implementation for mlx5 and hisi for this
new option to make it feature complete for the existing drivers in this
area.

Signed-off-by: Yishai Hadas <yishaih@nvidia.com>
Reviewed-by: Jason Gunthorpe <jgg@nvidia.com>
Reviewed-by: Longfang Liu <liulongfang@huawei.com>
Link: https://lore.kernel.org/r/20221106174630.25909-2-yishaih@nvidia.com
Signed-off-by: Alex Williamson <alex.williamson@redhat.com>

show more ...

hisi_acc_vfio_pci: Update some log and comment formats

1. Modify some annotation information formats to keep the
entire driver annotation format consistent.
2. Modify some log description formats to

hisi_acc_vfio_pci: Update some log and comment formats

1. Modify some annotation information formats to keep the
entire driver annotation format consistent.
2. Modify some log description formats to be consistent with
the format of the entire driver log.

Reviewed-by: Jason Gunthorpe <jgg@nvidia.com>
Signed-off-by: Longfang Liu <liulongfang@huawei.com>
Link: https://lore.kernel.org/r/20220926093332.28824-6-liulongfang@huawei.com
Signed-off-by: Alex Williamson <alex.williamson@redhat.com>

show more ...

hisi_acc_vfio_pci: Remove useless function parameter

Remove unused function parameters for vf_qm_fun_reset() and
ensure the device is enabled before the reset operation
is performed.

Reviewed-by: J

hisi_acc_vfio_pci: Remove useless function parameter

Remove unused function parameters for vf_qm_fun_reset() and
ensure the device is enabled before the reset operation
is performed.

Reviewed-by: Jason Gunthorpe <jgg@nvidia.com>
Signed-off-by: Longfang Liu <liulongfang@huawei.com>
Link: https://lore.kernel.org/r/20220926093332.28824-4-liulongfang@huawei.com
Signed-off-by: Alex Williamson <alex.williamson@redhat.com>

show more ...

hisi_acc_vfio_pci: Fix device data address combination problem

The queue address of the accelerator device should be combined into
a dma address in a way of combining the low and high bits.
The prev

hisi_acc_vfio_pci: Fix device data address combination problem

The queue address of the accelerator device should be combined into
a dma address in a way of combining the low and high bits.
The previous combination is wrong and needs to be modified.

Reviewed-by: Jason Gunthorpe <jgg@nvidia.com>
Signed-off-by: Longfang Liu <liulongfang@huawei.com>
Link: https://lore.kernel.org/r/20220926093332.28824-3-liulongfang@huawei.com
Signed-off-by: Alex Williamson <alex.williamson@redhat.com>

show more ...

hisi_acc_vfio_pci: Fixes error return code issue

During the process of compatibility and matching of live migration
device information, if the isolation status of the two devices is
inconsistent, th

hisi_acc_vfio_pci: Fixes error return code issue

During the process of compatibility and matching of live migration
device information, if the isolation status of the two devices is
inconsistent, the live migration needs to be exited.

The current driver does not return the error code correctly and
needs to be fixed.

Reviewed-by: Shameer Kolothum <shameerali.kolothum.thodi@huawei.com>
Reviewed-by: Jason Gunthorpe <jgg@nvidia.com>
Signed-off-by: Longfang Liu <liulongfang@huawei.com>
Link: https://lore.kernel.org/r/20220926093332.28824-2-liulongfang@huawei.com
Signed-off-by: Alex Williamson <alex.williamson@redhat.com>

show more ...

vfio/hisi_acc: Use the new device life cycle helpers

Tidy up @probe so all migration specific initialization logic is moved
to migration specific @init callback.

Remove vfio_pci_core_{un}init_devic

vfio/hisi_acc: Use the new device life cycle helpers

Tidy up @probe so all migration specific initialization logic is moved
to migration specific @init callback.

Remove vfio_pci_core_{un}init_device() given no user now.

Signed-off-by: Yi Liu <yi.l.liu@intel.com>
Signed-off-by: Kevin Tian <kevin.tian@intel.com>
Reviewed-by: Jason Gunthorpe <jgg@nvidia.com>
Reviewed-by: Shameer Kolothum <shameerali.kolothum.thodi@huawei.com>
Link: https://lore.kernel.org/r/20220921104401.38898-5-kevin.tian@intel.com
Signed-off-by: Alex Williamson <alex.williamson@redhat.com>

show more ...

hisi_acc_vfio_pci: Correct the function prefix for hssi_acc_drvdata()

Commit 91be0bd6c6cf("vfio/pci: Have all VFIO PCI drivers store the
vfio_pci_core_device in drvdata") introduced a helper functio

hisi_acc_vfio_pci: Correct the function prefix for hssi_acc_drvdata()

Commit 91be0bd6c6cf("vfio/pci: Have all VFIO PCI drivers store the
vfio_pci_core_device in drvdata") introduced a helper function to
retrieve the drvdata but used "hssi" instead of "hisi" for the
function prefix. Correct that and also while at it, moved the
function a bit down so that it's close to other hisi_ prefixed
functions.

No functional changes.

Signed-off-by: Shameer Kolothum <shameerali.kolothum.thodi@huawei.com>
Reviewed-by: Kevin Tian <kevin.tian@intel.com>
Reviewed-by: Jason Gunthorpe <jgg@nvidia.com>
Link: https://lore.kernel.org/r/20220831085943.993-1-shameerali.kolothum.thodi@huawei.com
Signed-off-by: Alex Williamson <alex.williamson@redhat.com>

show more ...

vfio: Split migration ops from main device ops

vfio core checks whether the driver sets some migration op (e.g.
set_state/get_state) and accordingly calls its op.

However, currently mlx5 driver set

vfio: Split migration ops from main device ops

vfio core checks whether the driver sets some migration op (e.g.
set_state/get_state) and accordingly calls its op.

However, currently mlx5 driver sets the above ops without regards to its
migration caps.

This might lead to unexpected usage/Oops if user space may call to the
above ops even if the driver doesn't support migration. As for example,
the migration state_mutex is not initialized in that case.

The cleanest way to manage that seems to split the migration ops from
the main device ops, this will let the driver setting them separately
from the main ops when it's applicable.

As part of that, validate ops construction on registration and include a
check for VFIO_MIGRATION_STOP_COPY since the uAPI claims it must be set
in migration_flags.

HISI driver was changed as well to match this scheme.

This scheme may enable down the road to come with some extra group of
ops (e.g. DMA log) that can be set without regards to the other options
based on driver caps.

Fixes: 6fadb021266d ("vfio/mlx5: Implement vfio_pci driver for mlx5 devices")
Reviewed-by: Kevin Tian <kevin.tian@intel.com>
Signed-off-by: Yishai Hadas <yishaih@nvidia.com>
Link: https://lore.kernel.org/r/20220628155910.171454-3-yishaih@nvidia.com
Signed-off-by: Alex Williamson <alex.williamson@redhat.com>

show more ...

vfio/pci: Add driver_managed_dma to the new vfio_pci drivers

When the iommu series adding driver_managed_dma was rebased it missed that
new VFIO drivers were added and did not update them too.

With

vfio/pci: Add driver_managed_dma to the new vfio_pci drivers

When the iommu series adding driver_managed_dma was rebased it missed that
new VFIO drivers were added and did not update them too.

Without this vfio will claim the groups are not viable.

Add driver_managed_dma to mlx5 and hisi.

Fixes: 70693f470848 ("vfio: Set DMA ownership for VFIO devices")
Reported-by: Yishai Hadas <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>
Link: https://lore.kernel.org/r/0-v1-f9dfa642fab0+2b3-vfio_managed_dma_jgg@nvidia.com
Signed-off-by: Alex Williamson <alex.williamson@redhat.com>

show more ...

vfio/pci: Have all VFIO PCI drivers store the vfio_pci_core_device in drvdata

Having a consistent pointer in the drvdata will allow the next patch to
make use of the drvdata from some of the core co

vfio/pci: Have all VFIO PCI drivers store the vfio_pci_core_device in drvdata

Having a consistent pointer in the drvdata will allow the next patch to
make use of the drvdata from some of the core code helpers.

Use a WARN_ON inside vfio_pci_core_register_device() to detect drivers
that miss this.

Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
Link: https://lore.kernel.org/r/1-v4-c841817a0349+8f-vfio_get_from_dev_jgg@nvidia.com
Signed-off-by: Alex Williamson <alex.williamson@redhat.com>

show more ...

hisi_acc_vfio_pci: Use its own PCI reset_done error handler

Register private handler for pci_error_handlers.reset_done and update
state accordingly.

Reviewed-by: Jason Gunthorpe <jgg@nvidia.com>
Re

hisi_acc_vfio_pci: Use its own PCI reset_done error handler

Register private handler for pci_error_handlers.reset_done and update
state accordingly.

Reviewed-by: Jason Gunthorpe <jgg@nvidia.com>
Reviewed-by: Longfang Liu <liulongfang@huawei.com>
Signed-off-by: Shameer Kolothum <shameerali.kolothum.thodi@huawei.com>
Link: https://lore.kernel.org/r/20220308184902.2242-10-shameerali.kolothum.thodi@huawei.com
Signed-off-by: Alex Williamson <alex.williamson@redhat.com>

show more ...

hisi_acc_vfio_pci: Add support for VFIO live migration

VMs assigned with HiSilicon ACC VF devices can now perform live migration
if the VF devices are bind to the hisi_acc_vfio_pci driver.

Just lik

hisi_acc_vfio_pci: Add support for VFIO live migration

VMs assigned with HiSilicon ACC VF devices can now perform live migration
if the VF devices are bind to the hisi_acc_vfio_pci driver.

Just like ACC PF/VF drivers this VFIO driver also make use of the HiSilicon
QM interface. QM stands for Queue Management which is a generic IP used by
ACC devices. It provides a generic PCIe interface for the CPU and the ACC
devices to share a group of queues.

QM integrated into an accelerator provides queue management service.
Queues can be assigned to PF and VFs, and queues can be controlled by
unified mailboxes and doorbells.

The QM driver (drivers/crypto/hisilicon/qm.c) provides generic
interfaces to ACC drivers to manage the QM.

Signed-off-by: Longfang Liu <liulongfang@huawei.com>
Reviewed-by: Jason Gunthorpe <jgg@nvidia.com>
Signed-off-by: Shameer Kolothum <shameerali.kolothum.thodi@huawei.com>
Link: https://lore.kernel.org/r/20220308184902.2242-9-shameerali.kolothum.thodi@huawei.com
Reviewed-by: Kevin Tian <kevin.tian@intel.com>
Signed-off-by: Alex Williamson <alex.williamson@redhat.com>

show more ...

hisi_acc_vfio_pci: Restrict access to VF dev BAR2 migration region

HiSilicon ACC VF device BAR2 region consists of both functional register
space and migration control register space. Unnecessarily 

hisi_acc_vfio_pci: Restrict access to VF dev BAR2 migration region

HiSilicon ACC VF device BAR2 region consists of both functional register
space and migration control register space. Unnecessarily exposing the
migration BAR region to the Guest has the potential to prevent/corrupt
the Guest migration.

Hence, introduce a separate struct vfio_device_ops for migration support
which will override the ioctl/read/write/mmap methods to hide the
migration region and limit the Guest access only to the functional
register space.

This will be used in subsequent patches when we add migration support
to the driver.

Please note that it is OK to export the entire VF BAR if migration is
not supported or required as this cannot affect the PF configurations.

Reviewed-by: Longfang Liu <liulongfang@huawei.com>
Signed-off-by: Shameer Kolothum <shameerali.kolothum.thodi@huawei.com>
Reviewed-by: Kevin Tian <kevin.tian@intel.com>
Link: https://lore.kernel.org/r/20220308184902.2242-6-shameerali.kolothum.thodi@huawei.com
Signed-off-by: Alex Williamson <alex.williamson@redhat.com>

show more ...

hisi_acc_vfio_pci: add new vfio_pci driver for HiSilicon ACC devices

Add a vendor-specific vfio_pci driver for HiSilicon ACC devices.
This will be extended in subsequent patches to add support for V

hisi_acc_vfio_pci: add new vfio_pci driver for HiSilicon ACC devices

Add a vendor-specific vfio_pci driver for HiSilicon ACC devices.
This will be extended in subsequent patches to add support for VFIO
live migration feature.

Signed-off-by: Shameer Kolothum <shameerali.kolothum.thodi@huawei.com>
Reviewed-by: Kevin Tian <kevin.tian@intel.com>
Link: https://lore.kernel.org/r/20220308184902.2242-5-shameerali.kolothum.thodi@huawei.com
Signed-off-by: Alex Williamson <alex.williamson@redhat.com>

show more ...