vfio/nvgrace-gpu: Add vfio pci variant module for grace hopper

NVIDIA's upcoming Grace Hopper Superchip provides a PCI-like device
for the on-chip GPU that is the logical OS representation of the
internal proprietary chip-to-chip cache coherent interconnect.

The device is peculiar compared to a real PCI device in that whilst
there is a real 64b PCI BAR1 (comprising region 2 & region 3) on the
device, it is not used to access device memory once the faster
chip-to-chip interconnect is initialized (occurs at the time of host
system boot). The device memory is accessed instead using the chip-to-chip
interconnect that is exposed as a contiguous physically addressable
region on the host. This device memory aperture can be obtained from host
ACPI table using device_property_read_u64(), according to the FW
specification. Since the device memory is cache coherent with the CPU,
it can be mmap into the user VMA with a cacheable mapping using
remap_pfn_range() and used like a regular RAM. The device memory
is not added to the host kernel, but mapped directly as this reduces
memory wastage due to struct pages.

There is also a requirement of a minimum reserved 1G uncached region
(termed as resmem) to support the Multi-Instance GPU (MIG) feature [1].
This is to work around a HW defect. Based on [2], the requisite properties
(uncached, unaligned access) can be achieved through a VM mapping (S1)
of NORMAL_NC and host (S2) mapping with MemAttr[2:0]=0b101. To provide
a different non-cached property to the reserved 1G region, it needs to
be carved out from the device memory and mapped as a separate region
in Qemu VMA with pgprot_writecombine(). pgprot_writecombine() sets the
Qemu VMA page properties (pgprot) as NORMAL_NC.

Provide a VFIO PCI variant driver that adapts the unique device memory
representation into a more standard PCI representation facing userspace.

The variant driver exposes these two regions - the non-cached reserved
(resmem) and the cached rest of the device memory (termed as usemem) as
separate VFIO 64b BAR regions. This is divergent from the baremetal
approach, where the device memory is exposed as a device memory region.
The decision for a different approach was taken in view of the fact that
it would necessiate additional code in Qemu to discover and insert those
regions in the VM IPA, along with the additional VM ACPI DSDT changes to
communicate the device memory region IPA to the VM workloads. Moreover,
this behavior would have to be added to a variety of emulators (beyond
top of tree Qemu) out there desiring grace hopper support.

Since the device implements 64-bit BAR0, the VFIO PCI variant driver
maps the uncached carved out region to the next available PCI BAR (i.e.
comprising of region 2 and 3). The cached device memory aperture is
assigned BAR region 4 and 5. Qemu will then naturally generate a PCI
device in the VM with the uncached aperture reported as BAR2 region,
the cacheable as BAR4. The variant driver provides emulation for these
fake BARs' PCI config space offset registers.

The hardware ensures that the system does not crash when the memory
is accessed with the memory enable turned off. It synthesis ~0 reads
and dropped writes on such access. So there is no need to support the
disablement/enablement of BAR through PCI_COMMAND config space register.

The memory layout on the host looks like the following:
devmem (memlength)
|--------------------------------------------------|
|-------------cached------------------------|--NC--|
| |
usemem.memphys resmem.memphys

PCI BARs need to be aligned to the power-of-2, but the actual memory on the
device may not. A read or write access to the physical address from the
last device PFN up to the next power-of-2 aligned physical address
results in reading ~0 and dropped writes. Note that the GPU device
driver [6] is capable of knowing the exact device memory size through
separate means. The device memory size is primarily kept in the system
ACPI tables for use by the VFIO PCI variant module.

Note that the usemem memory is added by the VM Nvidia device driver [5]
to the VM kernel as memblocks. Hence make the usable memory size memblock
(MEMBLK_SIZE) aligned. This is a hardwired ABI value between the GPU FW and
VFIO driver. The VM device driver make use of the same value for its
calculation to determine USEMEM size.

Currently there is no provision in KVM for a S2 mapping with
MemAttr[2:0]=0b101, but there is an ongoing effort to provide the same [3].
As previously mentioned, resmem is mapped pgprot_writecombine(), that
sets the Qemu VMA page properties (pgprot) as NORMAL_NC. Using the
proposed changes in [3] and [4], KVM marks the region with
MemAttr[2:0]=0b101 in S2.

If the device memory properties are not present, the driver registers the
vfio-pci-core function pointers. Since there are no ACPI memory properties
generated for the VM, the variant driver inside the VM will only use
the vfio-pci-core ops and hence try to map the BARs as non cached. This
is not a problem as the CPUs have FWB enabled which blocks the VM
mapping's ability to override the cacheability set by the host mapping.

This goes along with a qemu series [6] to provides the necessary
implementation of the Grace Hopper Superchip firmware specification so
that the guest operating system can see the correct ACPI modeling for
the coherent GPU device. Verified with the CUDA workload in the VM.

[1] https://www.nvidia.com/en-in/technologies/multi-instance-gpu/
[2] section D8.5.5 of https://developer.arm.com/documentation/ddi0487/latest/
[3] https://lore.kernel.org/all/20240211174705.31992-1-ankita@nvidia.com/
[4] https://lore.kernel.org/all/20230907181459.18145-2-ankita@nvidia.com/
[5] https://github.com/NVIDIA/open-gpu-kernel-modules
[6] https://lore.kernel.org/all/20231203060245.31593-1-ankita@nvidia.com/

Reviewed-by: Kevin Tian <kevin.tian@intel.com>
Reviewed-by: Yishai Hadas <yishaih@nvidia.com>
Reviewed-by: Zhi Wang <zhi.wang.linux@gmail.com>
Signed-off-by: Aniket Agashe <aniketa@nvidia.com>
Signed-off-by: Ankit Agrawal <ankita@nvidia.com>
Link: https://lore.kernel.org/r/20240220115055.23546-4-ankita@nvidia.com
Signed-off-by: Alex Williamson <alex.williamson@redhat.com>

vfio/virtio: Introduce a vfio driver over virtio devices

Introduce a vfio driver over virtio devices to support the legacy
interface functionality for VFs.

Background, from the virtio spec [1].
--------------------------------------------------------------------
In some systems, there is a need to support a virtio legacy driver with
a device that does not directly support the legacy interface. In such
scenarios, a group owner device can provide the legacy interface
functionality for the group member devices. The driver of the owner
device can then access the legacy interface of a member device on behalf
of the legacy member device driver.

For example, with the SR-IOV group type, group members (VFs) can not
present the legacy interface in an I/O BAR in BAR0 as expected by the
legacy pci driver. If the legacy driver is running inside a virtual
machine, the hypervisor executing the virtual machine can present a
virtual device with an I/O BAR in BAR0. The hypervisor intercepts the
legacy driver accesses to this I/O BAR and forwards them to the group
owner device (PF) using group administration commands.
--------------------------------------------------------------------

Specifically, this driver adds support for a virtio-net VF to be exposed
as a transitional device to a guest driver and allows the legacy IO BAR
functionality on top.

This allows a VM which uses a legacy virtio-net driver in the guest to
work transparently over a VF which its driver in the host is that new
driver.

The driver can be extended easily to support some other types of virtio
devices (e.g virtio-blk), by adding in a few places the specific type
properties as was done for virtio-net.

For now, only the virtio-net use case was tested and as such we introduce
the support only for such a device.

Practically,
Upon probing a VF for a virtio-net device, in case its PF supports
legacy access over the virtio admin commands and the VF doesn't have BAR
0, we set some specific 'vfio_device_ops' to be able to simulate in SW a
transitional device with I/O BAR in BAR 0.

The existence of the simulated I/O bar is reported later on by
overwriting the VFIO_DEVICE_GET_REGION_INFO command and the device
exposes itself as a transitional device by overwriting some properties
upon reading its config space.

Once we report the existence of I/O BAR as BAR 0 a legacy driver in the
guest may use it via read/write calls according to the virtio
specification.

Any read/write towards the control parts of the BAR will be captured by
the new driver and will be translated into admin commands towards the
device.

In addition, any data path read/write access (i.e. virtio driver
notifications) will be captured by the driver and forwarded to the
physical BAR which its properties were supplied by the admin command
VIRTIO_ADMIN_CMD_LEGACY_NOTIFY_INFO upon the probing/init flow.

With that code in place a legacy driver in the guest has the look and
feel as if having a transitional device with legacy support for both its
control and data path flows.

[1]
https://github.com/oasis-tcs/virtio-spec/commit/03c2d32e5093ca9f2a17797242fbef88efe94b8c

Reviewed-by: Jason Gunthorpe <jgg@nvidia.com>
Reviewed-by: Kevin Tian <kevin.tian@intel.com>
Signed-off-by: Yishai Hadas <yishaih@nvidia.com>
Acked-by: Michael S. Tsirkin <mst@redhat.com>
Link: https://lore.kernel.org/r/20231219093247.170936-10-yishaih@nvidia.com
Signed-off-by: Alex Williamson <alex.williamson@redhat.com>

vfio/pds: Initial support for pds VFIO driver

This is the initial framework for the new pds-vfio-pci device driver.
This does the very basics of registering the PDS PCI device and
configuring it as a VFIO PCI device.

With this change, the VF device can be bound to the pds-vfio-pci driver
on the host and presented to the VM as an ethernet VF.

Signed-off-by: Brett Creeley <brett.creeley@amd.com>
Signed-off-by: Shannon Nelson <shannon.nelson@amd.com>
Reviewed-by: Simon Horman <horms@kernel.org>
Reviewed-by: Kevin Tian <kevin.tian@intel.com>
Reviewed-by: Shameer Kolothum <shameerali.kolothum.thodi@huawei.com>
Reviewed-by: Jason Gunthorpe <jgg@nvidia.com>
Link: https://lore.kernel.org/r/20230807205755.29579-3-brett.creeley@amd.com
Signed-off-by: Alex Williamson <alex.williamson@redhat.com>

vfio/pci: introduce CONFIG_VFIO_PCI_ZDEV_KVM

The current contents of vfio-pci-zdev are today only useful in a KVM
environment; let's tie everything currently under vfio-pci-zdev to
this Kconfig statement and require KVM in this case, reducing complexity
(e.g. symbol lookups).

Signed-off-by: Matthew Rosato <mjrosato@linux.ibm.com>
Acked-by: Alex Williamson <alex.williamson@redhat.com>
Reviewed-by: Pierre Morel <pmorel@linux.ibm.com>
Link: https://lore.kernel.org/r/20220606203325.110625-11-mjrosato@linux.ibm.com
Signed-off-by: Christian Borntraeger <borntraeger@linux.ibm.com>

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>

vfio/mlx5: Implement vfio_pci driver for mlx5 devices

This patch adds support for vfio_pci driver for mlx5 devices.

It uses vfio_pci_core to register to the VFIO subsystem and then
implements the mlx5 specific logic in the migration area.

The migration implementation follows the definition from uapi/vfio.h and
uses the mlx5 VF->PF command channel to achieve it.

This patch implements the suspend/resume flows.

Link: https://lore.kernel.org/all/20220224142024.147653-14-yishaih@nvidia.com
Reviewed-by: Alex Williamson <alex.williamson@redhat.com>
Signed-off-by: Yishai Hadas <yishaih@nvidia.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
Signed-off-by: Leon Romanovsky <leonro@nvidia.com>

vfio/pci: Introduce vfio_pci_core.ko

Now that vfio_pci has been split into two source modules, one focusing on
the "struct pci_driver" (vfio_pci.c) and a toolbox library of code
(vfio_pci_core.c), complete the split and move them into two different
kernel modules.

As before vfio_pci.ko continues to present the same interface under sysfs
and this change will have no functional impact.

Splitting into another module and adding exports allows creating new HW
specific VFIO PCI drivers that can implement device specific
functionality, such as VFIO migration interfaces or specialized device
requirements.

Signed-off-by: Max Gurtovoy <mgurtovoy@nvidia.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Yishai Hadas <yishaih@nvidia.com>
Link: https://lore.kernel.org/r/20210826103912.128972-14-yishaih@nvidia.com
Signed-off-by: Alex Williamson <alex.williamson@redhat.com>

vfio/pci: Split the pci_driver code out of vfio_pci_core.c

Split the vfio_pci driver into two logical parts, the 'struct
pci_driver' (vfio_pci.c) which implements "Generic VFIO support for any
PCI device" and a library of code (vfio_pci_core.c) that helps
implementing a struct vfio_device on top of a PCI device.

vfio_pci.ko continues to present the same interface under sysfs and this
change should have no functional impact.

Following patches will turn vfio_pci and vfio_pci_core into a separate
module.

This is a preparation for allowing another module to provide the
pci_driver and allow that module to customize how VFIO is setup, inject
its own operations, and easily extend vendor specific functionality.

At this point the vfio_pci_core still contains a lot of vfio_pci
functionality mixed into it. Following patches will move more of the
large scale items out, but another cleanup series will be needed to get
everything.

Signed-off-by: Max Gurtovoy <mgurtovoy@nvidia.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Yishai Hadas <yishaih@nvidia.com>
Link: https://lore.kernel.org/r/20210826103912.128972-7-yishaih@nvidia.com
Signed-off-by: Alex Williamson <alex.williamson@redhat.com>

vfio/pci: Rename vfio_pci.c to vfio_pci_core.c

This is a preparation patch for separating the vfio_pci driver to a
subsystem driver and a generic pci driver. This patch doesn't change any
logic.

Signed-off-by: Max Gurtovoy <mgurtovoy@nvidia.com>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Yishai Hadas <yishaih@nvidia.com>
Link: https://lore.kernel.org/r/20210826103912.128972-2-yishaih@nvidia.com
Signed-off-by: Alex Williamson <alex.williamson@redhat.com>

vfio/pci: remove vfio_pci_nvlink2

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>

vfio/pci: remove CONFIG_VFIO_PCI_ZDEV from Kconfig

In case we're running on s390 system always expose the capabilities for
configuration of zPCI devices. In case we're running on different
platform, continue as usual.

Signed-off-by: Max Gurtovoy <mgurtovoy@nvidia.com>
Reviewed-by: Matthew Rosato <mjrosato@linux.ibm.com>
Reviewed-by: Cornelia Huck <cohuck@redhat.com>
Signed-off-by: Alex Williamson <alex.williamson@redhat.com>

vfio-pci/zdev: Add zPCI capabilities to VFIO_DEVICE_GET_INFO

Define a new configuration entry VFIO_PCI_ZDEV for VFIO/PCI.

When this s390-only feature is configured we add capabilities to the
VFIO_DEVICE_GET_INFO ioctl that describe features of the associated
zPCI device and its underlying hardware.

This patch is based on work previously done by Pierre Morel.

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>

treewide: Add SPDX license identifier - Makefile/Kconfig

Add SPDX license identifiers to all Make/Kconfig files which:

- Have no license information of any form

These files fall under the project license, GPL v2 only. The resulting SPDX
license identifier is:

GPL-2.0-only

Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

vfio_pci: Add NVIDIA GV100GL [Tesla V100 SXM2] subdriver

POWER9 Witherspoon machines come with 4 or 6 V100 GPUs which are not
pluggable PCIe devices but still have PCIe links which are used
for config space and MMIO. In addition to that the GPUs have 6 NVLinks
which are connected to other GPUs and the POWER9 CPU. POWER9 chips
have a special unit on a die called an NPU which is an NVLink2 host bus
adapter with p2p connections to 2 to 3 GPUs, 3 or 2 NVLinks to each.
These systems also support ATS (address translation services) which is
a part of the NVLink2 protocol. Such GPUs also share on-board RAM
(16GB or 32GB) to the system via the same NVLink2 so a CPU has
cache-coherent access to a GPU RAM.

This exports GPU RAM to the userspace as a new VFIO device region. This
preregisters the new memory as device memory as it might be used for DMA.
This inserts pfns from the fault handler as the GPU memory is not onlined
until the vendor driver is loaded and trained the NVLinks so doing this
earlier causes low level errors which we fence in the firmware so
it does not hurt the host system but still better be avoided; for the same
reason this does not map GPU RAM into the host kernel (usual thing for
emulated access otherwise).

This exports an ATSD (Address Translation Shootdown) register of NPU which
allows TLB invalidations inside GPU for an operating system. The register
conveniently occupies a single 64k page. It is also presented to
the userspace as a new VFIO device region. One NPU has 8 ATSD registers,
each of them can be used for TLB invalidation in a GPU linked to this NPU.
This allocates one ATSD register per an NVLink bridge allowing passing
up to 6 registers. Due to the host firmware bug (just recently fixed),
only 1 ATSD register per NPU was actually advertised to the host system
so this passes that alone register via the first NVLink bridge device in
the group which is still enough as QEMU collects them all back and
presents to the guest via vPHB to mimic the emulated NPU PHB on the host.

In order to provide the userspace with the information about GPU-to-NVLink
connections, this exports an additional capability called "tgt"
(which is an abbreviated host system bus address). The "tgt" property
tells the GPU its own system address and allows the guest driver to
conglomerate the routing information so each GPU knows how to get directly
to the other GPUs.

For ATS to work, the nest MMU (an NVIDIA block in a P9 CPU) needs to
know LPID (a logical partition ID or a KVM guest hardware ID in other
words) and PID (a memory context ID of a userspace process, not to be
confused with a linux pid). This assigns a GPU to LPID in the NPU and
this is why this adds a listener for KVM on an IOMMU group. A PID comes
via NVLink from a GPU and NPU uses a PID wildcard to pass it through.

This requires coherent memory and ATSD to be available on the host as
the GPU vendor only supports configurations with both features enabled
and other configurations are known not to work. Because of this and
because of the ways the features are advertised to the host system
(which is a device tree with very platform specific properties),
this requires enabled POWERNV platform.

The V100 GPUs do not advertise any of these capabilities via the config
space and there are more than just one device ID so this relies on
the platform to tell whether these GPUs have special abilities such as
NVLinks.

Signed-off-by: Alexey Kardashevskiy <aik@ozlabs.ru>
Acked-by: Alex Williamson <alex.williamson@redhat.com>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>

vfio/pci: Intel IGD OpRegion support

This is the first consumer of vfio device specific resource support,
providing read-only access to the OpRegion for Intel graphics devices.

Signed-off-by: Alex Williamson <alex.williamson@redhat.com>

vfio: initialize the virqfd workqueue in VFIO generic code

Now we have finally completely decoupled virqfd from VFIO_PCI. We can
initialize it from the VFIO generic code, in order to safely use it from
multiple independent VFIO bus drivers.

Signed-off-by: Antonios Motakis <a.motakis@virtualopensystems.com>
Signed-off-by: Baptiste Reynal <b.reynal@virtualopensystems.com>
Reviewed-by: Eric Auger <eric.auger@linaro.org>
Tested-by: Eric Auger <eric.auger@linaro.org>
Signed-off-by: Alex Williamson <alex.williamson@redhat.com>

vfio: move eventfd support code for VFIO_PCI to a separate file

The virqfd functionality that is used by VFIO_PCI to implement interrupt
masking and unmasking via an eventfd, is generic enough and can be reused
by another driver. Move it to a separate file in order to allow the code
to be shared.

Signed-off-by: Antonios Motakis <a.motakis@virtualopensystems.com>
Signed-off-by: Baptiste Reynal <b.reynal@virtualopensystems.com>
Reviewed-by: Eric Auger <eric.auger@linaro.org>
Tested-by: Eric Auger <eric.auger@linaro.org>
Signed-off-by: Alex Williamson <alex.williamson@redhat.com>

vfio: Add PCI device driver

Add PCI device support for VFIO. PCI devices expose regions
for accessing config space, I/O port space, and MMIO areas
of the device. PCI config access is virtualized in the kernel,
allowing us to ensure the integrity of the system, by preventing
various accesses while reducing duplicate support across various
userspace drivers. I/O port supports read/write access while
MMIO also supports mmap of sufficiently sized regions. Support
for INTx, MSI, and MSI-X interrupts are provided using eventfds to
userspace.

Signed-off-by: Alex Williamson <alex.williamson@redhat.com>