There is really no way to safely give a user full access to a DMA
capable device without an IOMMU to protect the host system. There is
also no way to provide DMA translation, for use cases such as device
assignment to virtual machines. However, there are still those users
that want userspace drivers even under those conditions. The UIO
driver exists for this use case, but does not provide the degree of
device access and programming that VFIO has. In an effort to avoid
code duplication, this introduces a No-IOMMU mode for VFIO.
This mode requires building VFIO with CONFIG_VFIO_NOIOMMU and enabling
the "enable_unsafe_noiommu_mode" option on the vfio driver. This
should make it very clear that this mode is not safe. Additionally,
CAP_SYS_RAWIO privileges are necessary to work with groups and
containers using this mode. Groups making use of this support are
named /dev/vfio/noiommu-$GROUP and can only make use of the special
VFIO_NOIOMMU_IOMMU for the container. Use of this mode, specifically
binding a device without a native IOMMU group to a VFIO bus driver
will taint the kernel and should therefore not be considered
supported. This patch includes no-iommu support for the vfio-pci bus
driver only.
Signed-off-by: Alex Williamson <alex.williamson@redhat.com>
Acked-by: Michael S. Tsirkin <mst@redhat.com>
The vfio_device_get_from_name() function might return a
non-NULL pointer, when called with a device name that is not
found in the list. This causes undefined behavior, in my
case calling an invalid function pointer later on:
kernel tried to execute NX-protected page - exploit attempt? (uid: 0)
BUG: unable to handle kernel paging request at ffff8800cb3ddc08
[...]
Call Trace:
[<ffffffffa03bd733>] ? vfio_group_fops_unl_ioctl+0x253/0x410 [vfio]
[<ffffffff811efc4d>] do_vfs_ioctl+0x2cd/0x4c0
[<ffffffff811f9657>] ? __fget+0x77/0xb0
[<ffffffff811efeb9>] SyS_ioctl+0x79/0x90
[<ffffffff81001bb0>] ? syscall_return_slowpath+0x50/0x130
[<ffffffff8167f776>] entry_SYSCALL_64_fastpath+0x16/0x75
Fix the issue by returning NULL when there is no device with
the requested name in the list.
Cc: stable@vger.kernel.org # v4.2+
Fixes: 4bc94d5dc9 ("vfio: Fix lockdep issue")
Signed-off-by: Joerg Roedel <jroedel@suse.de>
Signed-off-by: Alex Williamson <alex.williamson@redhat.com>
This patch introduces a module that registers and implements a low-level
reset function for the AMD XGBE device.
it performs the following actions:
- reset the PHY
- disable auto-negotiation
- disable & clear auto-negotiation IRQ
- soft-reset the MAC
Those tiny pieces of code are inherited from the native xgbe driver.
Signed-off-by: Eric Auger <eric.auger@linaro.org>
Reviewed-by: Arnd Bergmann <arnd@arndb.de>
Signed-off-by: Alex Williamson <alex.williamson@redhat.com>
In the current code the vfio_platform_region is copied on the stack.
As a consequence the ioaddr address is not iounmapped in the vfio
platform driver (vfio_platform_regions_cleanup). The patch uses the
pointer to the region instead.
Signed-off-by: Eric Auger <eric.auger@linaro.org>
Signed-off-by: Alex Williamson <alex.williamson@redhat.com>
It might be helpful for the end-user to check the device reset
function was found by the vfio platform reset framework.
Lets store a pointer to the struct device in vfio_platform_device
and trace when the reset function is called or not found.
Signed-off-by: Eric Auger <eric.auger@linaro.org>
Signed-off-by: Alex Williamson <alex.williamson@redhat.com>
Remove the static lookup table and use the dynamic list of registered
reset functions instead. Also load the reset module through its alias.
The reset struct module pointer is stored in vfio_platform_device.
We also remove the useless struct device pointer parameter in
vfio_platform_get_reset.
This patch fixes the issue related to the usage of __symbol_get, which
besides from being moot, prevented compilation with CONFIG_MODULES
disabled.
Also usage of MODULE_ALIAS makes possible to add a new reset module
without needing to update the framework. This was suggested by Arnd.
Signed-off-by: Eric Auger <eric.auger@linaro.org>
Reported-by: Arnd Bergmann <arnd@arndb.de>
Reviewed-by: Arnd Bergmann <arnd@arndb.de>
Signed-off-by: Alex Williamson <alex.williamson@redhat.com>
Let's retrieve the compatibility string on probe and store it
in the vfio_platform_device struct
Signed-off-by: Eric Auger <eric.auger@linaro.org>
Signed-off-by: Alex Williamson <alex.williamson@redhat.com>
This patch adds the reset function registration/unregistration.
This is handled through the module_vfio_reset_handler macro. This
latter also defines a MODULE_ALIAS which simplifies the load from
vfio-platform.
Signed-off-by: Eric Auger <eric.auger@linaro.org>
Reviewed-by: Arnd Bergmann <arnd@arndb.de>
Signed-off-by: Alex Williamson <alex.williamson@redhat.com>
The module_vfio_reset_handler macro
- define a module alias
- implement module init/exit function which respectively registers
and unregisters the reset function.
Signed-off-by: Eric Auger <eric.auger@linaro.org>
Reviewed-by: Arnd Bergmann <arnd@arndb.de>
Signed-off-by: Alex Williamson <alex.williamson@redhat.com>
In preparation for subsequent changes in reset function lookup,
lets introduce a dynamic list of reset combos (compat string,
reset module, reset function). The list can be populated/voided with
vfio_platform_register/unregister_reset. Those are not yet used in
this patch.
Signed-off-by: Eric Auger <eric.auger@linaro.org>
Reviewed-by: Arnd Bergmann <arnd@arndb.de>
Signed-off-by: Alex Williamson <alex.williamson@redhat.com>
To prepare for vfio platform reset rework let's build
vfio_platform_common.c and vfio_platform_irq.c in a separate
module from vfio-platform and vfio-amba. This makes possible
to have separate module inits and works around a race between
platform driver init and vfio reset module init: that way we
make sure symbols exported by base are available when vfio-platform
driver gets probed.
The open/release being implemented in the base module, the ref
count is applied to the parent module instead.
Signed-off-by: Eric Auger <eric.auger@linaro.org>
Suggested-by: Arnd Bergmann <arnd@arndb.de>
Reviewed-by: Arnd Bergmann <arnd@arndb.de>
Signed-off-by: Alex Williamson <alex.williamson@redhat.com>
vfio_platform_{read,write}_mmio() call ioremap_nocache() to map
a region of io memory, which they store in struct vfio_platform_region to
be eventually re-used, or unmapped by vfio_platform_regions_cleanup().
These functions receive a copy of their struct vfio_platform_region
argument on the stack - so these mapped areas are always allocated, and
always leaked.
Pass this argument as a pointer instead.
Fixes: 6e3f264560 "vfio/platform: read and write support for the device fd"
Signed-off-by: James Morse <james.morse@arm.com>
Acked-by: Baptiste Reynal <b.reynal@virtualopensystems.com>
Tested-by: Baptiste Reynal <b.reynal@virtualopensystems.com>
Signed-off-by: Alex Williamson <alex.williamson@redhat.com>
Current vfio_pgsize_bitmap code hides the supported IOMMU page
sizes smaller than PAGE_SIZE. As a result, in case the IOMMU
does not support PAGE_SIZE page, the alignment check on map/unmap
is done with larger page sizes, if any. This can fail although
mapping could be done with pages smaller than PAGE_SIZE.
This patch modifies vfio_pgsize_bitmap implementation so that,
in case the IOMMU supports page sizes smaller than PAGE_SIZE
we pretend PAGE_SIZE is supported and hide sub-PAGE_SIZE sizes.
That way the user will be able to map/unmap buffers whose size/
start address is aligned with PAGE_SIZE. Pinning code uses that
granularity while iommu driver can use the sub-PAGE_SIZE size
to map the buffer.
Signed-off-by: Eric Auger <eric.auger@linaro.org>
Acked-by: Will Deacon <will.deacon@arm.com>
Signed-off-by: Alex Williamson <alex.williamson@redhat.com>
The vfio platform driver currently sets the IRQ_NOAUTOEN before
doing the request_irq to properly handle the user masking. However
it does not clear it when de-assigning the IRQ. This brings issues
when loading the native driver again which may not explicitly enable
the IRQ. This problem was observed with xgbe driver.
Signed-off-by: Eric Auger <eric.auger@linaro.org>
Signed-off-by: Alex Williamson <alex.williamson@redhat.com>
The PCI VPD capability operates on a set of window registers in PCI
config space. Writing to the address register triggers either a read
or write, depending on the setting of the PCI_VPD_ADDR_F bit within
the address register. The data register provides either the source
for writes or the target for reads.
This model is susceptible to being broken by concurrent access, for
which the kernel has adopted a set of access functions to serialize
these registers. Additionally, commits like 932c435cab ("PCI: Add
dev_flags bit to access VPD through function 0") and 7aa6ca4d39
("PCI: Add VPD function 0 quirk for Intel Ethernet devices") indicate
that VPD registers can be shared between functions on multifunction
devices creating dependencies between otherwise independent devices.
Fortunately it's quite easy to emulate the VPD registers, simply
storing copies of the address and data registers in memory and
triggering a VPD read or write on writes to the address register.
This allows vfio users to avoid seeing spurious register changes from
accesses on other devices and enables the use of shared quirks in the
host kernel. We can theoretically still race with access through
sysfs, but the window of opportunity is much smaller.
Signed-off-by: Alex Williamson <alex.williamson@redhat.com>
Acked-by: Mark Rustad <mark.d.rustad@intel.com>
When determining whether a group is viable, we already allow devices
bound to pcieport. Generalize this to include any PCI bridge device.
Signed-off-by: Alex Williamson <alex.williamson@redhat.com>
When we open a device file descriptor, we currently have the
following:
vfio_group_get_device_fd()
mutex_lock(&group->device_lock);
open()
...
if (ret)
release()
If we hit that error case, we call the backend driver release path,
which for vfio-pci looks like this:
vfio_pci_release()
vfio_pci_disable()
vfio_pci_try_bus_reset()
vfio_pci_get_devs()
vfio_device_get_from_dev()
vfio_group_get_device()
mutex_lock(&group->device_lock);
Whoops, we've stumbled back onto group.device_lock and created a
deadlock. There's a low likelihood of ever seeing this play out, but
obviously it needs to be fixed. To do that we can use a reference to
the vfio_device for vfio_group_get_device_fd() rather than holding the
lock. There was a loop in this function, theoretically allowing
multiple devices with the same name, but in practice we don't expect
such a thing to happen and the code is already aborting from the loop
with break on any sort of error rather than continuing and only
parsing the first match anyway, so the loop was effectively unused
already.
Signed-off-by: Alex Williamson <alex.williamson@redhat.com>
Fixes: 20f300175a ("vfio/pci: Fix racy vfio_device_get_from_dev() call")
Reported-by: Joerg Roedel <joro@8bytes.org>
Tested-by: Joerg Roedel <jroedel@suse.de>
- Disable the 32-bit vdso when building LE, so we can build with a 64-bit only
toolchain.
- EEH fixes from Gavin & Richard.
- Enable the sys_kcmp syscall from Laurent.
- Sysfs control for fastsleep workaround from Shreyas.
- Expose OPAL events as an irq chip by Alistair.
- MSI ops moved to pci_controller_ops by Daniel.
- Fix for kernel to userspace backtraces for perf from Anton.
- Merge pseries and pseries_le defconfigs from Cyril.
- CXL in-kernel API from Mikey.
- OPAL prd driver from Jeremy.
- Fix for DSCR handling & tests from Anshuman.
- Powernv flash mtd driver from Cyril.
- Dynamic DMA Window support on powernv from Alexey.
- LLVM clang fixes & workarounds from Anton.
- Reworked version of the patch to abort syscalls when transactional.
- Fix the swap encoding to support 4TB, from Aneesh.
- Various fixes as usual.
- Freescale updates from Scott: Highlights include more 8xx optimizations, an
e6500 hugetlb optimization, QMan device tree nodes, t1024/t1023 support, and
various fixes and cleanup.
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1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=d9Wh
-----END PGP SIGNATURE-----
Merge tag 'powerpc-4.2-1' of git://git.kernel.org/pub/scm/linux/kernel/git/mpe/linux
Pull powerpc updates from Michael Ellerman:
- disable the 32-bit vdso when building LE, so we can build with a
64-bit only toolchain.
- EEH fixes from Gavin & Richard.
- enable the sys_kcmp syscall from Laurent.
- sysfs control for fastsleep workaround from Shreyas.
- expose OPAL events as an irq chip by Alistair.
- MSI ops moved to pci_controller_ops by Daniel.
- fix for kernel to userspace backtraces for perf from Anton.
- merge pseries and pseries_le defconfigs from Cyril.
- CXL in-kernel API from Mikey.
- OPAL prd driver from Jeremy.
- fix for DSCR handling & tests from Anshuman.
- Powernv flash mtd driver from Cyril.
- dynamic DMA Window support on powernv from Alexey.
- LLVM clang fixes & workarounds from Anton.
- reworked version of the patch to abort syscalls when transactional.
- fix the swap encoding to support 4TB, from Aneesh.
- various fixes as usual.
- Freescale updates from Scott: Highlights include more 8xx
optimizations, an e6500 hugetlb optimization, QMan device tree nodes,
t1024/t1023 support, and various fixes and cleanup.
* tag 'powerpc-4.2-1' of git://git.kernel.org/pub/scm/linux/kernel/git/mpe/linux: (180 commits)
cxl: Fix typo in debug print
cxl: Add CXL_KERNEL_API config option
powerpc/powernv: Fix wrong IOMMU table in pnv_ioda_setup_bus_dma()
powerpc/mm: Change the swap encoding in pte.
powerpc/mm: PTE_RPN_MAX is not used, remove the same
powerpc/tm: Abort syscalls in active transactions
powerpc/iommu/ioda2: Enable compile with IOV=on and IOMMU_API=off
powerpc/include: Add opal-prd to installed uapi headers
powerpc/powernv: fix construction of opal PRD messages
powerpc/powernv: Increase opal-irqchip initcall priority
powerpc: Make doorbell check preemption safe
powerpc/powernv: pnv_init_idle_states() should only run on powernv
macintosh/nvram: Remove as unused
powerpc: Don't use gcc specific options on clang
powerpc: Don't use -mno-strict-align on clang
powerpc: Only use -mtraceback=no, -mno-string and -msoft-float if toolchain supports it
powerpc: Only use -mabi=altivec if toolchain supports it
powerpc: Fix duplicate const clang warning in user access code
vfio: powerpc/spapr: Support Dynamic DMA windows
vfio: powerpc/spapr: Register memory and define IOMMU v2
...
This patch enables building VFIO platform and derivatives on ARM64.
Signed-off-by: Eric Auger <eric.auger@linaro.org>
Acked-by: Baptiste Reynal <b.reynal@virtualopensystems.com>
Tested-by: Baptiste Reynal <b.reynal@virtualopensystems.com>
Signed-off-by: Alex Williamson <alex.williamson@redhat.com>
This patch introduces a module that registers and implements a basic
reset function for the Calxeda xgmac device. This latter basically disables
interrupts and stops DMA transfers.
The reset function code is inherited from the native calxeda xgmac driver.
Signed-off-by: Eric Auger <eric.auger@linaro.org>
Acked-by: Baptiste Reynal <b.reynal@virtualopensystems.com>
Tested-by: Baptiste Reynal <b.reynal@virtualopensystems.com>
Signed-off-by: Alex Williamson <alex.williamson@redhat.com>
The reset function lookup happens on vfio-platform probe. The reset
module load is requested and a reference to the function symbol is
hold. The reference is released on vfio-platform remove.
Signed-off-by: Eric Auger <eric.auger@linaro.org>
Acked-by: Baptiste Reynal <b.reynal@virtualopensystems.com>
Tested-by: Baptiste Reynal <b.reynal@virtualopensystems.com>
Signed-off-by: Alex Williamson <alex.williamson@redhat.com>
A new reset callback is introduced. If this callback is populated,
the reset is invoked on device first open/last close or upon userspace
ioctl. The modality is exposed on VFIO_DEVICE_GET_INFO.
Signed-off-by: Eric Auger <eric.auger@linaro.org>
Acked-by: Baptiste Reynal <b.reynal@virtualopensystems.com>
Tested-by: Baptiste Reynal <b.reynal@virtualopensystems.com>
Signed-off-by: Alex Williamson <alex.williamson@redhat.com>
This patch introduces the vfio_platform_reset_combo struct that
stores all the information useful to handle the reset modality:
compat string, name of the reset function, name of the module that
implements the reset function. A lookup table of such structures
is added, currently void.
Signed-off-by: Eric Auger <eric.auger@linaro.org>
Acked-by: Baptiste Reynal <b.reynal@virtualopensystems.com>
Tested-by: Baptiste Reynal <b.reynal@virtualopensystems.com>
Signed-off-by: Alex Williamson <alex.williamson@redhat.com>
This adds create/remove window ioctls to create and remove DMA windows.
sPAPR defines a Dynamic DMA windows capability which allows
para-virtualized guests to create additional DMA windows on a PCI bus.
The existing linux kernels use this new window to map the entire guest
memory and switch to the direct DMA operations saving time on map/unmap
requests which would normally happen in a big amounts.
This adds 2 ioctl handlers - VFIO_IOMMU_SPAPR_TCE_CREATE and
VFIO_IOMMU_SPAPR_TCE_REMOVE - to create and remove windows.
Up to 2 windows are supported now by the hardware and by this driver.
This changes VFIO_IOMMU_SPAPR_TCE_GET_INFO handler to return additional
information such as a number of supported windows and maximum number
levels of TCE tables.
DDW is added as a capability, not as a SPAPR TCE IOMMU v2 unique feature
as we still want to support v2 on platforms which cannot do DDW for
the sake of TCE acceleration in KVM (coming soon).
Signed-off-by: Alexey Kardashevskiy <aik@ozlabs.ru>
[aw: for the vfio related changes]
Acked-by: Alex Williamson <alex.williamson@redhat.com>
Reviewed-by: David Gibson <david@gibson.dropbear.id.au>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
The existing implementation accounts the whole DMA window in
the locked_vm counter. This is going to be worse with multiple
containers and huge DMA windows. Also, real-time accounting would requite
additional tracking of accounted pages due to the page size difference -
IOMMU uses 4K pages and system uses 4K or 64K pages.
Another issue is that actual pages pinning/unpinning happens on every
DMA map/unmap request. This does not affect the performance much now as
we spend way too much time now on switching context between
guest/userspace/host but this will start to matter when we add in-kernel
DMA map/unmap acceleration.
This introduces a new IOMMU type for SPAPR - VFIO_SPAPR_TCE_v2_IOMMU.
New IOMMU deprecates VFIO_IOMMU_ENABLE/VFIO_IOMMU_DISABLE and introduces
2 new ioctls to register/unregister DMA memory -
VFIO_IOMMU_SPAPR_REGISTER_MEMORY and VFIO_IOMMU_SPAPR_UNREGISTER_MEMORY -
which receive user space address and size of a memory region which
needs to be pinned/unpinned and counted in locked_vm.
New IOMMU splits physical pages pinning and TCE table update
into 2 different operations. It requires:
1) guest pages to be registered first
2) consequent map/unmap requests to work only with pre-registered memory.
For the default single window case this means that the entire guest
(instead of 2GB) needs to be pinned before using VFIO.
When a huge DMA window is added, no additional pinning will be
required, otherwise it would be guest RAM + 2GB.
The new memory registration ioctls are not supported by
VFIO_SPAPR_TCE_IOMMU. Dynamic DMA window and in-kernel acceleration
will require memory to be preregistered in order to work.
The accounting is done per the user process.
This advertises v2 SPAPR TCE IOMMU and restricts what the userspace
can do with v1 or v2 IOMMUs.
In order to support memory pre-registration, we need a way to track
the use of every registered memory region and only allow unregistration
if a region is not in use anymore. So we need a way to tell from what
region the just cleared TCE was from.
This adds a userspace view of the TCE table into iommu_table struct.
It contains userspace address, one per TCE entry. The table is only
allocated when the ownership over an IOMMU group is taken which means
it is only used from outside of the powernv code (such as VFIO).
As v2 IOMMU supports IODA2 and pre-IODA2 IOMMUs (which do not support
DDW API), this creates a default DMA window for IODA2 for consistency.
Signed-off-by: Alexey Kardashevskiy <aik@ozlabs.ru>
[aw: for the vfio related changes]
Acked-by: Alex Williamson <alex.williamson@redhat.com>
Reviewed-by: David Gibson <david@gibson.dropbear.id.au>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Before the IOMMU user (VFIO) would take control over the IOMMU table
belonging to a specific IOMMU group. This approach did not allow sharing
tables between IOMMU groups attached to the same container.
This introduces a new IOMMU ownership flavour when the user can not
just control the existing IOMMU table but remove/create tables on demand.
If an IOMMU implements take/release_ownership() callbacks, this lets
the user have full control over the IOMMU group. When the ownership
is taken, the platform code removes all the windows so the caller must
create them.
Before returning the ownership back to the platform code, VFIO
unprograms and removes all the tables it created.
This changes IODA2's onwership handler to remove the existing table
rather than manipulating with the existing one. From now on,
iommu_take_ownership() and iommu_release_ownership() are only called
from the vfio_iommu_spapr_tce driver.
Old-style ownership is still supported allowing VFIO to run on older
P5IOC2 and IODA IO controllers.
No change in userspace-visible behaviour is expected. Since it recreates
TCE tables on each ownership change, related kernel traces will appear
more often.
This adds a pnv_pci_ioda2_setup_default_config() which is called
when PE is being configured at boot time and when the ownership is
passed from VFIO to the platform code.
Signed-off-by: Alexey Kardashevskiy <aik@ozlabs.ru>
[aw: for the vfio related changes]
Acked-by: Alex Williamson <alex.williamson@redhat.com>
Reviewed-by: David Gibson <david@gibson.dropbear.id.au>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
This extends iommu_table_group_ops by a set of callbacks to support
dynamic DMA windows management.
create_table() creates a TCE table with specific parameters.
it receives iommu_table_group to know nodeid in order to allocate
TCE table memory closer to the PHB. The exact format of allocated
multi-level table might be also specific to the PHB model (not
the case now though).
This callback calculated the DMA window offset on a PCI bus from @num
and stores it in a just created table.
set_window() sets the window at specified TVT index + @num on PHB.
unset_window() unsets the window from specified TVT.
This adds a free() callback to iommu_table_ops to free the memory
(potentially a tree of tables) allocated for the TCE table.
create_table() and free() are supposed to be called once per
VFIO container and set_window()/unset_window() are supposed to be
called for every group in a container.
This adds IOMMU capabilities to iommu_table_group such as default
32bit window parameters and others. This makes use of new values in
vfio_iommu_spapr_tce. IODA1/P5IOC2 do not support DDW so they do not
advertise pagemasks to the userspace.
Signed-off-by: Alexey Kardashevskiy <aik@ozlabs.ru>
Acked-by: Alex Williamson <alex.williamson@redhat.com>
Reviewed-by: David Gibson <david@gibson.dropbear.id.au>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
At the moment writing new TCE value to the IOMMU table fails with EBUSY
if there is a valid entry already. However PAPR specification allows
the guest to write new TCE value without clearing it first.
Another problem this patch is addressing is the use of pool locks for
external IOMMU users such as VFIO. The pool locks are to protect
DMA page allocator rather than entries and since the host kernel does
not control what pages are in use, there is no point in pool locks and
exchange()+put_page(oldtce) is sufficient to avoid possible races.
This adds an exchange() callback to iommu_table_ops which does the same
thing as set() plus it returns replaced TCE and DMA direction so
the caller can release the pages afterwards. The exchange() receives
a physical address unlike set() which receives linear mapping address;
and returns a physical address as the clear() does.
This implements exchange() for P5IOC2/IODA/IODA2. This adds a requirement
for a platform to have exchange() implemented in order to support VFIO.
This replaces iommu_tce_build() and iommu_clear_tce() with
a single iommu_tce_xchg().
This makes sure that TCE permission bits are not set in TCE passed to
IOMMU API as those are to be calculated by platform code from
DMA direction.
This moves SetPageDirty() to the IOMMU code to make it work for both
VFIO ioctl interface in in-kernel TCE acceleration (when it becomes
available later).
Signed-off-by: Alexey Kardashevskiy <aik@ozlabs.ru>
[aw: for the vfio related changes]
Acked-by: Alex Williamson <alex.williamson@redhat.com>
Reviewed-by: David Gibson <david@gibson.dropbear.id.au>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
This adds tce_iommu_take_ownership() and tce_iommu_release_ownership
which call in a loop iommu_take_ownership()/iommu_release_ownership()
for every table on the group. As there is just one now, no change in
behaviour is expected.
At the moment the iommu_table struct has a set_bypass() which enables/
disables DMA bypass on IODA2 PHB. This is exposed to POWERPC IOMMU code
which calls this callback when external IOMMU users such as VFIO are
about to get over a PHB.
The set_bypass() callback is not really an iommu_table function but
IOMMU/PE function. This introduces a iommu_table_group_ops struct and
adds take_ownership()/release_ownership() callbacks to it which are
called when an external user takes/releases control over the IOMMU.
This replaces set_bypass() with ownership callbacks as it is not
necessarily just bypass enabling, it can be something else/more
so let's give it more generic name.
The callbacks is implemented for IODA2 only. Other platforms (P5IOC2,
IODA1) will use the old iommu_take_ownership/iommu_release_ownership API.
The following patches will replace iommu_take_ownership/
iommu_release_ownership calls in IODA2 with full IOMMU table release/
create.
As we here and touching bypass control, this removes
pnv_pci_ioda2_setup_bypass_pe() as it does not do much
more compared to pnv_pci_ioda2_set_bypass. This moves tce_bypass_base
initialization to pnv_pci_ioda2_setup_dma_pe.
Signed-off-by: Alexey Kardashevskiy <aik@ozlabs.ru>
[aw: for the vfio related changes]
Acked-by: Alex Williamson <alex.williamson@redhat.com>
Reviewed-by: Gavin Shan <gwshan@linux.vnet.ibm.com>
Reviewed-by: David Gibson <david@gibson.dropbear.id.au>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
So far one TCE table could only be used by one IOMMU group. However
IODA2 hardware allows programming the same TCE table address to
multiple PE allowing sharing tables.
This replaces a single pointer to a group in a iommu_table struct
with a linked list of groups which provides the way of invalidating
TCE cache for every PE when an actual TCE table is updated. This adds
pnv_pci_link_table_and_group() and pnv_pci_unlink_table_and_group()
helpers to manage the list. However without VFIO, it is still going
to be a single IOMMU group per iommu_table.
This changes iommu_add_device() to add a device to a first group
from the group list of a table as it is only called from the platform
init code or PCI bus notifier and at these moments there is only
one group per table.
This does not change TCE invalidation code to loop through all
attached groups in order to simplify this patch and because
it is not really needed in most cases. IODA2 is fixed in a later
patch.
This should cause no behavioural change.
Signed-off-by: Alexey Kardashevskiy <aik@ozlabs.ru>
[aw: for the vfio related changes]
Acked-by: Alex Williamson <alex.williamson@redhat.com>
Reviewed-by: Gavin Shan <gwshan@linux.vnet.ibm.com>
Reviewed-by: David Gibson <david@gibson.dropbear.id.au>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Modern IBM POWERPC systems support multiple (currently two) TCE tables
per IOMMU group (a.k.a. PE). This adds a iommu_table_group container
for TCE tables. Right now just one table is supported.
This defines iommu_table_group struct which stores pointers to
iommu_group and iommu_table(s). This replaces iommu_table with
iommu_table_group where iommu_table was used to identify a group:
- iommu_register_group();
- iommudata of generic iommu_group;
This removes @data from iommu_table as it_table_group provides
same access to pnv_ioda_pe.
For IODA, instead of embedding iommu_table, the new iommu_table_group
keeps pointers to those. The iommu_table structs are allocated
dynamically.
For P5IOC2, both iommu_table_group and iommu_table are embedded into
PE struct. As there is no EEH and SRIOV support for P5IOC2,
iommu_free_table() should not be called on iommu_table struct pointers
so we can keep it embedded in pnv_phb::p5ioc2.
For pSeries, this replaces multiple calls of kzalloc_node() with a new
iommu_pseries_alloc_group() helper and stores the table group struct
pointer into the pci_dn struct. For release, a iommu_table_free_group()
helper is added.
This moves iommu_table struct allocation from SR-IOV code to
the generic DMA initialization code in pnv_pci_ioda_setup_dma_pe and
pnv_pci_ioda2_setup_dma_pe as this is where DMA is actually initialized.
This change is here because those lines had to be changed anyway.
This should cause no behavioural change.
Signed-off-by: Alexey Kardashevskiy <aik@ozlabs.ru>
[aw: for the vfio related changes]
Acked-by: Alex Williamson <alex.williamson@redhat.com>
Reviewed-by: David Gibson <david@gibson.dropbear.id.au>
Reviewed-by: Gavin Shan <gwshan@linux.vnet.ibm.com>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
This is to make extended ownership and multiple groups support patches
simpler for review.
This should cause no behavioural change.
Signed-off-by: Alexey Kardashevskiy <aik@ozlabs.ru>
[aw: for the vfio related changes]
Acked-by: Alex Williamson <alex.williamson@redhat.com>
Reviewed-by: David Gibson <david@gibson.dropbear.id.au>
Reviewed-by: Gavin Shan <gwshan@linux.vnet.ibm.com>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
This is a pretty mechanical patch to make next patches simpler.
New tce_iommu_unuse_page() helper does put_page() now but it might skip
that after the memory registering patch applied.
As we are here, this removes unnecessary checks for a value returned
by pfn_to_page() as it cannot possibly return NULL.
This moves tce_iommu_disable() later to let tce_iommu_clear() know if
the container has been enabled because if it has not been, then
put_page() must not be called on TCEs from the TCE table. This situation
is not yet possible but it will after KVM acceleration patchset is
applied.
This changes code to work with physical addresses rather than linear
mapping addresses for better code readability. Following patches will
add an xchg() callback for an IOMMU table which will accept/return
physical addresses (unlike current tce_build()) which will eliminate
redundant conversions.
Signed-off-by: Alexey Kardashevskiy <aik@ozlabs.ru>
[aw: for the vfio related changes]
Acked-by: Alex Williamson <alex.williamson@redhat.com>
Reviewed-by: David Gibson <david@gibson.dropbear.id.au>
Reviewed-by: Gavin Shan <gwshan@linux.vnet.ibm.com>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
At the moment DMA map/unmap requests are handled irrespective to
the container's state. This allows the user space to pin memory which
it might not be allowed to pin.
This adds checks to MAP/UNMAP that the container is enabled, otherwise
-EPERM is returned.
Signed-off-by: Alexey Kardashevskiy <aik@ozlabs.ru>
[aw: for the vfio related changes]
Acked-by: Alex Williamson <alex.williamson@redhat.com>
Reviewed-by: David Gibson <david@gibson.dropbear.id.au>
Reviewed-by: Gavin Shan <gwshan@linux.vnet.ibm.com>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
There moves locked pages accounting to helpers.
Later they will be reused for Dynamic DMA windows (DDW).
This reworks debug messages to show the current value and the limit.
This stores the locked pages number in the container so when unlocking
the iommu table pointer won't be needed. This does not have an effect
now but it will with the multiple tables per container as then we will
allow attaching/detaching groups on fly and we may end up having
a container with no group attached but with the counter incremented.
While we are here, update the comment explaining why RLIMIT_MEMLOCK
might be required to be bigger than the guest RAM. This also prints
pid of the current process in pr_warn/pr_debug.
Signed-off-by: Alexey Kardashevskiy <aik@ozlabs.ru>
[aw: for the vfio related changes]
Acked-by: Alex Williamson <alex.williamson@redhat.com>
Reviewed-by: David Gibson <david@gibson.dropbear.id.au>
Reviewed-by: Gavin Shan <gwshan@linux.vnet.ibm.com>
Reviewed-by: David Gibson <david@gibson.dropbear.id.au>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
This makes use of the it_page_size from the iommu_table struct
as page size can differ.
This replaces missing IOMMU_PAGE_SHIFT macro in commented debug code
as recently introduced IOMMU_PAGE_XXX macros do not include
IOMMU_PAGE_SHIFT.
Signed-off-by: Alexey Kardashevskiy <aik@ozlabs.ru>
Reviewed-by: David Gibson <david@gibson.dropbear.id.au>
[aw: for the vfio related changes]
Acked-by: Alex Williamson <alex.williamson@redhat.com>
Reviewed-by: Gavin Shan <gwshan@linux.vnet.ibm.com>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
This checks that the TCE table page size is not bigger that the size of
a page we just pinned and going to put its physical address to the table.
Otherwise the hardware gets unwanted access to physical memory between
the end of the actual page and the end of the aligned up TCE page.
Since compound_order() and compound_head() work correctly on non-huge
pages, there is no need for additional check whether the page is huge.
Signed-off-by: Alexey Kardashevskiy <aik@ozlabs.ru>
[aw: for the vfio related changes]
Acked-by: Alex Williamson <alex.williamson@redhat.com>
Reviewed-by: David Gibson <david@gibson.dropbear.id.au>
Reviewed-by: Gavin Shan <gwshan@linux.vnet.ibm.com>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
This moves page pinning (get_user_pages_fast()/put_page()) code out of
the platform IOMMU code and puts it to VFIO IOMMU driver where it belongs
to as the platform code does not deal with page pinning.
This makes iommu_take_ownership()/iommu_release_ownership() deal with
the IOMMU table bitmap only.
This removes page unpinning from iommu_take_ownership() as the actual
TCE table might contain garbage and doing put_page() on it is undefined
behaviour.
Besides the last part, the rest of the patch is mechanical.
Signed-off-by: Alexey Kardashevskiy <aik@ozlabs.ru>
[aw: for the vfio related changes]
Acked-by: Alex Williamson <alex.williamson@redhat.com>
Reviewed-by: David Gibson <david@gibson.dropbear.id.au>
Reviewed-by: Gavin Shan <gwshan@linux.vnet.ibm.com>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Testing the driver for a PCI device is racy, it can be all but
complete in the release path and still report the driver as ours.
Therefore we can't trust drvdata to be valid. This race can sometimes
be seen when one port of a multifunction device is being unbound from
the vfio-pci driver while another function is being released by the
user and attempting a bus reset. The device in the remove path is
found as a dependent device for the bus reset of the release path
device, the driver is still set to vfio-pci, but the drvdata has
already been cleared, resulting in a null pointer dereference.
To resolve this, fix vfio_device_get_from_dev() to not take the
dev_get_drvdata() shortcut and instead traverse through the
iommu_group, vfio_group, vfio_device path to get a reference we
can trust. Once we have that reference, we know the device isn't
in transition and we can test to make sure the driver is still what
we expect, so that we don't interfere with devices we don't own.
Signed-off-by: Alex Williamson <alex.williamson@redhat.com>
The ARM SMMUv3 driver is compatible with the notion of a type-1 IOMMU in
VFIO.
This patch allows VFIO_IOMMU_TYPE1 to be selected if ARM_SMMU_V3=y.
Signed-off-by: Will Deacon <will.deacon@arm.com>
Acked-by: Alex Williamson <alex.williamson@redhat.com>
Signed-off-by: Joerg Roedel <jroedel@suse.de>
The patch adds one more EEH sub-command (VFIO_EEH_PE_INJECT_ERR)
to inject the specified EEH error, which is represented by
(struct vfio_eeh_pe_err), to the indicated PE for testing purpose.
Signed-off-by: Gavin Shan <gwshan@linux.vnet.ibm.com>
Reviewed-by: David Gibson <david@gibson.dropbear.id.au>
Acked-by: Alex Williamson <alex.williamson@redhat.com>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Commit 13060b64b8 ("vfio: Add and use device request op for vfio
bus drivers") incorrectly makes use of an interruptible timeout.
When interrupted, the signal remains pending resulting in subsequent
timeouts occurring instantly. This makes the loop spin at a much
higher rate than intended.
Instead of making this completely non-interruptible, we can change
this into a sort of interruptible-once behavior and use the "once"
to log debug information. The driver API doesn't allow us to abort
and return an error code.
Signed-off-by: Alex Williamson <alex.williamson@redhat.com>
Fixes: 13060b64b8
Cc: stable@vger.kernel.org # v4.0
Log some clues indicating whether the user is receiving device
request interfaces or not listening. This can help indicate why a
driver unbind is blocked or explain why QEMU automatically unplugged
a device from the VM.
Signed-off-by: Alex Williamson <alex.williamson@redhat.com>
We can save some power by putting devices that are bound to vfio-pci
but not in use by the user in the D3hot power state. Devices get
woken into D0 when opened by the user. Resets return the device to
D0, so we need to re-apply the low power state after a bus reset.
It's tempting to try to use D3cold, but we have no reason to inhibit
hotplug of idle devices and we might get into a loop of having the
device disappear before we have a chance to try to use it.
A new module parameter allows this feature to be disabled if there are
devices that misbehave as a result of this change.
Signed-off-by: Alex Williamson <alex.williamson@redhat.com>
As indicated in the comment, this is not entirely uncommon and
causes user concern for no reason.
Signed-off-by: Alex Williamson <alex.williamson@redhat.com>
This copies the same support from pci-stub for exactly the same
purpose, enabling a set of PCI IDs to be automatically added to the
driver's dynamic ID table at module load time. The code here is
pretty simple and both vfio-pci and pci-stub are fairly unique in
being meta drivers, capable of attaching to any device, so there's no
attempt made to generalize the code into pci-core.
Signed-off-by: Alex Williamson <alex.williamson@redhat.com>
If VFIO VGA access is disabled for the user, either by CONFIG option
or module parameter, we can often opt-out of VGA arbitration. We can
do this when PCI bridge control of VGA routing is possible. This
means that we must have a parent bridge and there must only be a
single VGA device below that bridge. Fortunately this is the typical
case for discrete GPUs.
Doing this allows us to minimize the impact of additional GPUs, in
terms of VGA arbitration, when they are only used via vfio-pci for
non-VGA applications.
Signed-off-by: Alex Williamson <alex.williamson@redhat.com>
Add a module option so that we don't require a CONFIG change and
kernel rebuild to disable VGA support. Not only can VGA support be
troublesome in itself, but by disabling it we can reduce the impact
to host devices by doing a VGA arbitration opt-out.
Signed-off-by: Alex Williamson <alex.williamson@redhat.com>
