We use a VFIOContainer to associate an AddressSpace to one or more
VFIOGroups. The VFIOContainer represents the DMA context for that
AdressSpace for those VFIOGroups and is synchronized to changes in
that AddressSpace via a MemoryListener. For IOMMU backed devices,
maintaining the DMA context for a VFIOGroup generally involves
pinning a host virtual address in order to create a stable host
physical address and then mapping a translation from the associated
guest physical address to that host physical address into the IOMMU.
While the above maintains the VFIOContainer synchronized to the QEMU
memory API of the VM, memory ballooning occurs outside of that API.
Inflating the memory balloon (ie. cooperatively capturing pages from
the guest for use by the host) simply uses MADV_DONTNEED to "zap"
pages from QEMU's host virtual address space. The page pinning and
IOMMU mapping above remains in place, negating the host's ability to
reuse the page, but the host virtual to host physical mapping of the
page is invalidated outside of QEMU's memory API.
When the balloon is later deflated, attempting to cooperatively
return pages to the guest, the page is simply freed by the guest
balloon driver, allowing it to be used in the guest and incurring a
page fault when that occurs. The page fault maps a new host physical
page backing the existing host virtual address, meanwhile the
VFIOContainer still maintains the translation to the original host
physical address. At this point the guest vCPU and any assigned
devices will map different host physical addresses to the same guest
physical address. Badness.
The IOMMU typically does not have page level granularity with which
it can track this mapping without also incurring inefficiencies in
using page size mappings throughout. MMU notifiers in the host
kernel also provide indicators for invalidating the mapping on
balloon inflation, not for updating the mapping when the balloon is
deflated. For these reasons we assume a default behavior that the
mapping of each VFIOGroup into the VFIOContainer is incompatible
with memory ballooning and increment the balloon inhibitor to match
the attached VFIOGroups.
Reviewed-by: Peter Xu <peterx@redhat.com>
Signed-off-by: Alex Williamson <alex.williamson@redhat.com>
Add support for multiple IOMMU indexes to the IOMMU notifier APIs.
When initializing a notifier with iommu_notifier_init(), the caller
must pass the IOMMU index that it is interested in. When a change
happens, the IOMMU implementation must pass
memory_region_notify_iommu() the IOMMU index that has changed and
that notifiers must be called for.
IOMMUs which support only a single index don't need to change.
Callers which only really support working with IOMMUs with a single
index can use the result of passing MEMTXATTRS_UNSPECIFIED to
memory_region_iommu_attrs_to_index().
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Reviewed-by: Alex Bennée <alex.bennee@linaro.org>
Message-id: 20180604152941.20374-3-peter.maydell@linaro.org
As part of plumbing MemTxAttrs down to the IOMMU translate method,
add MemTxAttrs as an argument to address_space_translate()
and address_space_translate_cached(). Callers either have an
attrs value to hand, or don't care and can use MEMTXATTRS_UNSPECIFIED.
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
Reviewed-by: Alex Bennée <alex.bennee@linaro.org>
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Message-id: 20180521140402.23318-4-peter.maydell@linaro.org
Commit 567b5b309a ("vfio/pci: Relax DMA map errors for MMIO regions")
added an error message if a passed memory section address or size
is not aligned to the page size and thus cannot be DMA mapped.
This patch fixes the trace by printing the region name and the
memory region section offset within the address space (instead of
offset_within_region).
We also turn the error_report into a trace event. Indeed, In some
cases, the traces can be confusing to non expert end-users and
let think the use case does not work (whereas it works as before).
This is the case where a BAR is successively mapped at different
GPAs and its sections are not compatible with dma map. The listener
is called several times and traces are issued for each intermediate
mapping. The end-user cannot easily match those GPAs against the
final GPA output by lscpi. So let's keep those information to
informed users. In mid term, the plan is to advise the user about
BAR relocation relevance.
Fixes: 567b5b309a ("vfio/pci: Relax DMA map errors for MMIO regions")
Signed-off-by: Eric Auger <eric.auger@redhat.com>
Reviewed-by: Philippe Mathieu-Daudé <f4bug@amsat.org>
Reviewed-by: Alexey Kardashevskiy <aik@ozlabs.ru>
Signed-off-by: Alex Williamson <alex.williamson@redhat.com>
At the moment we unconditionally avoid mapping MSIX data of a BAR and
emulate MSIX table in QEMU. However it is 1) not always necessary as
a platform may provide a paravirt interface for MSIX configuration;
2) can affect the speed of MMIO access by emulating them in QEMU when
frequently accessed registers share same system page with MSIX data,
this is particularly a problem for systems with the page size bigger
than 4KB.
A new capability - VFIO_REGION_INFO_CAP_MSIX_MAPPABLE - has been added
to the kernel [1] which tells the userspace that mapping of the MSIX data
is possible now. This makes use of it so from now on QEMU tries mapping
the entire BAR as a whole and emulate MSIX on top of that.
[1] https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/commit/?id=a32295c612c57990d17fb0f41e7134394b2f35f6
Signed-off-by: Alexey Kardashevskiy <aik@ozlabs.ru>
Reviewed-by: David Gibson <david@gibson.dropbear.id.au>
Signed-off-by: Alex Williamson <alex.williamson@redhat.com>
At the moment if vfio_memory_listener is registered in the system memory
address space, it maps/unmaps every RAM memory region for DMA.
It expects system page size aligned memory sections so vfio_dma_map
would not fail and so far this has been the case. A mapping failure
would be fatal. A side effect of such behavior is that some MMIO pages
would not be mapped silently.
However we are going to change MSIX BAR handling so we will end having
non-aligned sections in vfio_memory_listener (more details is in
the next patch) and vfio_dma_map will exit QEMU.
In order to avoid fatal failures on what previously was not a failure and
was just silently ignored, this checks the section alignment to
the smallest supported IOMMU page size and prints an error if not aligned;
it also prints an error if vfio_dma_map failed despite the page size check.
Both errors are not fatal; only MMIO RAM regions are checked
(aka "RAM device" regions).
If the amount of errors printed is overwhelming, the MSIX relocation
could be used to avoid excessive error output.
This is unlikely to cause any behavioral change.
Signed-off-by: Alexey Kardashevskiy <aik@ozlabs.ru>
[aw: Fix Int128 bit ops]
Signed-off-by: Alex Williamson <alex.williamson@redhat.com>
After next patch, listener unregister will need the container to be
alive. Let's move this unregister phase to be before unset container,
since that operation will free the backend container in kernel,
otherwise we'll get these after next patch:
qemu-system-x86_64: VFIO_UNMAP_DMA: -22
qemu-system-x86_64: vfio_dma_unmap(0x559bf53a4590, 0x0, 0xa0000) = -22 (Invalid argument)
Signed-off-by: Peter Xu <peterx@redhat.com>
Message-Id: <20180122060244.29368-4-peterx@redhat.com>
Reviewed-by: Paolo Bonzini <pbonzini@redhat.com>
Acked-by: Alex Williamson <alex.williamson@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
There is already @hostwin in vfio_listener_region_add() so there is no
point in having the other one.
Fixes: 2e4109de8e ("vfio/spapr: Create DMA window dynamically (SPAPR IOMMU v2)")
Signed-off-by: Alexey Kardashevskiy <aik@ozlabs.ru>
Signed-off-by: Alex Williamson <alex.williamson@redhat.com>
In order to enable TCE operations support in KVM, we have to inform
the KVM about VFIO groups being attached to specific LIOBNs. The KVM
already knows about VFIO groups, the only bit missing is which
in-kernel TCE table (the one with user visible TCEs) should update
the attached broups. There is an KVM_DEV_VFIO_GROUP_SET_SPAPR_TCE
attribute of the VFIO KVM device which receives a groupfd/tablefd couple.
This uses a new memory_region_iommu_get_attr() helper to get the IOMMU fd
and calls KVM to establish the link.
As get_attr() is not implemented yet, this should cause no behavioural
change.
Signed-off-by: Alexey Kardashevskiy <aik@ozlabs.ru>
Acked-by: Paolo Bonzini <pbonzini@redhat.com>
Acked-by: David Gibson <david@gibson.dropbear.id.au>
Signed-off-by: Alex Williamson <alex.williamson@redhat.com>
The vfio_iommu_spapr_tce driver advertises kernel's support for
v1 and v2 IOMMU support, however it is not always possible to use
the requested IOMMU type. For example, a pseries host platform does not
support dynamic DMA windows so v2 cannot initialize and QEMU fails to
start.
This adds a fallback to the v1 IOMMU if v2 cannot be used.
Fixes: 318f67ce13 ("vfio: spapr: Add DMA memory preregistering (SPAPR IOMMU v2)")
Signed-off-by: Alexey Kardashevskiy <aik@ozlabs.ru>
Signed-off-by: Alex Williamson <alex.williamson@redhat.com>
The init of giommu_list and hostwin_list is missed during container
initialization.
Signed-off-by: Liu, Yi L <yi.l.liu@linux.intel.com>
Signed-off-by: Alex Williamson <alex.williamson@redhat.com>
Commit 8c37faa475 ("vfio-pci, ppc64/spapr: Reorder group-to-container
attaching") moved registration of groups with the vfio-kvm device from
vfio_get_group() to vfio_connect_container(), but it missed the case
where a group is attached to an existing container and takes an early
exit. Perhaps this is a less common case on ppc64/spapr, but on x86
(without viommu) all groups are connected to the same container and
thus only the first group gets registered with the vfio-kvm device.
This becomes a problem if we then hot-unplug the devices associated
with that first group and we end up with KVM being misinformed about
any vfio connections that might remain. Fix by including the call to
vfio_kvm_device_add_group() in this early exit path.
Fixes: 8c37faa475 ("vfio-pci, ppc64/spapr: Reorder group-to-container attaching")
Cc: qemu-stable@nongnu.org # qemu-2.10+
Reviewed-by: Alexey Kardashevskiy <aik@ozlabs.ru>
Reviewed-by: Peter Xu <peterx@redhat.com>
Tested-by: Peter Xu <peterx@redhat.com>
Reviewed-by: Eric Auger <eric.auger@redhat.com>
Tested-by: Eric Auger <eric.auger@redhat.com>
Signed-off-by: Alex Williamson <alex.williamson@redhat.com>
At the moment VFIO PCI device initialization works as follows:
vfio_realize
vfio_get_group
vfio_connect_container
register memory listeners (1)
update QEMU groups lists
vfio_kvm_device_add_group
Then (example for pseries) the machine reset hook triggers region_add()
for all regions where listeners from (1) are listening:
ppc_spapr_reset
spapr_phb_reset
spapr_tce_table_enable
memory_region_add_subregion
vfio_listener_region_add
vfio_spapr_create_window
This scheme works fine until we need to handle VFIO PCI device hotplug
and we want to enable PPC64/sPAPR in-kernel TCE acceleration on,
i.e. after PCI hotplug we need a place to call
ioctl(vfio_kvm_device_fd, KVM_DEV_VFIO_GROUP_SET_SPAPR_TCE).
Since the ioctl needs a LIOBN fd (from sPAPRTCETable) and a IOMMU group fd
(from VFIOGroup), vfio_listener_region_add() seems to be the only place
for this ioctl().
However this only works during boot time because the machine reset
happens strictly after all devices are finalized. When hotplug happens,
vfio_listener_region_add() is called when a memory listener is registered
but when this happens:
1. new group is not added to the container->group_list yet;
2. VFIO KVM device is unaware of the new IOMMU group.
This moves bits around to have all necessary VFIO infrastructure
in place for both initial startup and hotplug cases.
[aw: ie, register vfio groups with kvm prior to memory listener
registration such that kvm-vfio pseudo device ioctls are available
during the region_add callback]
Signed-off-by: Alexey Kardashevskiy <aik@ozlabs.ru>
Reviewed-by: David Gibson <david@gibson.dropbear.id.au>
Signed-off-by: Alex Williamson <alex.williamson@redhat.com>
This defines new QOM object - IOMMUMemoryRegion - with MemoryRegion
as a parent.
This moves IOMMU-related fields from MR to IOMMU MR. However to avoid
dymanic QOM casting in fast path (address_space_translate, etc),
this adds an @is_iommu boolean flag to MR and provides new helper to
do simple cast to IOMMU MR - memory_region_get_iommu. The flag
is set in the instance init callback. This defines
memory_region_is_iommu as memory_region_get_iommu()!=NULL.
This switches MemoryRegion to IOMMUMemoryRegion in most places except
the ones where MemoryRegion may be an alias.
Signed-off-by: Alexey Kardashevskiy <aik@ozlabs.ru>
Reviewed-by: David Gibson <david@gibson.dropbear.id.au>
Message-Id: <20170711035620.4232-2-aik@ozlabs.ru>
Acked-by: Cornelia Huck <cohuck@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
VFIOGroup.device_list is effectively our reference tracking mechanism
such that we can teardown a group when all of the device references
are removed. However, we also use this list from our machine reset
handler for processing resets that affect multiple devices. Generally
device removals are fully processed (exitfn + finalize) when this
reset handler is invoked, however if the removal is triggered via
another reset handler (piix4_reset->acpi_pcihp_reset) then the device
exitfn may run, but not finalize. In this case we hit asserts when
we start trying to access PCI helpers since much of the PCI state of
the device is released. To resolve this, add a pointer to the Object
DeviceState in our common base-device and skip non-realized devices
as we iterate.
Signed-off-by: Alex Williamson <alex.williamson@redhat.com>
We were always passing in that one as "false" to assume that's an read
operation, and we also assume that IOMMU translation would always have
that read permission. A better permission would be IOMMU_NONE since the
replay is after all not a real read operation, but just a page table
rebuilding process.
CC: David Gibson <david@gibson.dropbear.id.au>
CC: Paolo Bonzini <pbonzini@redhat.com>
Reviewed-by: David Gibson <david@gibson.dropbear.id.au>
Acked-by: Paolo Bonzini <pbonzini@redhat.com>
Signed-off-by: Peter Xu <peterx@redhat.com>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
Reviewed-by: Jason Wang <jasowang@redhat.com>
This patch enables 8-byte writes and reads to VFIO. Such implemention
is already done but it's missing the 'case' to handle such accesses in
both vfio_region_write and vfio_region_read and the MemoryRegionOps:
impl.max_access_size and impl.min_access_size.
After this patch, 8-byte writes such as:
qemu_mutex_lock locked mutex 0x10905ad8
vfio_region_write (0001:03:00.0:region1+0xc0, 0x4140c, 4)
vfio_region_write (0001:03:00.0:region1+0xc4, 0xa0000, 4)
qemu_mutex_unlock unlocked mutex 0x10905ad8
goes like this:
qemu_mutex_lock locked mutex 0x10905ad8
vfio_region_write (0001:03:00.0:region1+0xc0, 0xbfd0008, 8)
qemu_mutex_unlock unlocked mutex 0x10905ad8
Signed-off-by: Jose Ricardo Ziviani <joserz@linux.vnet.ibm.com>
Signed-off-by: Alex Williamson <alex.williamson@redhat.com>
Sets valid.max_access_size and valid.min_access_size to ensure safe
8-byte accesses to vfio. Today, 8-byte accesses are broken into pairs
of 4-byte calls that goes unprotected:
qemu_mutex_lock locked mutex 0x10905ad8
vfio_region_write (0001:03:00.0:region1+0xc0, 0x2020c, 4)
qemu_mutex_unlock unlocked mutex 0x10905ad8
qemu_mutex_lock locked mutex 0x10905ad8
vfio_region_write (0001:03:00.0:region1+0xc4, 0xa0000, 4)
qemu_mutex_unlock unlocked mutex 0x10905ad8
which occasionally leads to:
qemu_mutex_lock locked mutex 0x10905ad8
vfio_region_write (0001:03:00.0:region1+0xc0, 0x2030c, 4)
qemu_mutex_unlock unlocked mutex 0x10905ad8
qemu_mutex_lock locked mutex 0x10905ad8
vfio_region_write (0001:03:00.0:region1+0xc0, 0x1000c, 4)
qemu_mutex_unlock unlocked mutex 0x10905ad8
qemu_mutex_lock locked mutex 0x10905ad8
vfio_region_write (0001:03:00.0:region1+0xc4, 0xb0000, 4)
qemu_mutex_unlock unlocked mutex 0x10905ad8
qemu_mutex_lock locked mutex 0x10905ad8
vfio_region_write (0001:03:00.0:region1+0xc4, 0xa0000, 4)
qemu_mutex_unlock unlocked mutex 0x10905ad8
causing strange errors in guest OS. With this patch, such accesses
are protected by the same lock guard:
qemu_mutex_lock locked mutex 0x10905ad8
vfio_region_write (0001:03:00.0:region1+0xc0, 0x2000c, 4)
vfio_region_write (0001:03:00.0:region1+0xc4, 0xb0000, 4)
qemu_mutex_unlock unlocked mutex 0x10905ad8
This happens because the 8-byte write should be broken into 4-byte
writes by memory.c:access_with_adjusted_size() in order to be under
the same lock. Today, it's done in exec.c:address_space_write_continue()
which was able to handle only 4 bytes due to a zero'ed
valid.max_access_size (see exec.c:memory_access_size()).
Signed-off-by: Jose Ricardo Ziviani <joserz@linux.vnet.ibm.com>
Signed-off-by: Alex Williamson <alex.williamson@redhat.com>
In this patch, IOMMUNotifier.{start|end} are introduced to store section
information for a specific notifier. When notification occurs, we not
only check the notification type (MAP|UNMAP), but also check whether the
notified iova range overlaps with the range of specific IOMMU notifier,
and skip those notifiers if not in the listened range.
When removing an region, we need to make sure we removed the correct
VFIOGuestIOMMU by checking the IOMMUNotifier.start address as well.
This patch is solving the problem that vfio-pci devices receive
duplicated UNMAP notification on x86 platform when vIOMMU is there. The
issue is that x86 IOMMU has a (0, 2^64-1) IOMMU region, which is
splitted by the (0xfee00000, 0xfeefffff) IRQ region. AFAIK
this (splitted IOMMU region) is only happening on x86.
This patch also helps vhost to leverage the new interface as well, so
that vhost won't get duplicated cache flushes. In that sense, it's an
slight performance improvement.
Suggested-by: David Gibson <david@gibson.dropbear.id.au>
Reviewed-by: Eric Auger <eric.auger@redhat.com>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Acked-by: Alex Williamson <alex.williamson@redhat.com>
Signed-off-by: Peter Xu <peterx@redhat.com>
Message-Id: <1491562755-23867-2-git-send-email-peterx@redhat.com>
[ehabkost: included extra vhost_iommu_region_del() change from Peter Xu]
Signed-off-by: Eduardo Habkost <ehabkost@redhat.com>
Linux vfio driver supports to do VFIO_IOMMU_UNMAP_DMA for a very big
region. This can be leveraged by QEMU IOMMU implementation to cleanup
existing page mappings for an entire iova address space (by notifying
with an IOTLB with extremely huge addr_mask). However current
vfio_iommu_map_notify() does not allow that. It make sure that all the
translated address in IOTLB is falling into RAM range.
The check makes sense, but it should only be a sensible checker for
mapping operations, and mean little for unmap operations.
This patch moves this check into map logic only, so that we'll get
faster unmap handling (no need to translate again), and also we can then
better support unmapping a very big region when it covers non-ram ranges
or even not-existing ranges.
Acked-by: Alex Williamson <alex.williamson@redhat.com>
Signed-off-by: Peter Xu <peterx@redhat.com>
Reviewed-by: David Gibson <david@gibson.dropbear.id.au>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
A cleanup for vfio_iommu_map_notify(). Now we will fetch vaddr even if
the operation is unmap, but it won't hurt much.
One thing to mention is that we need the RCU read lock to protect the
whole translation and map/unmap procedure.
Acked-by: Alex Williamson <alex.williamson@redhat.com>
Reviewed-by: David Gibson <david@gibson.dropbear.id.au>
Signed-off-by: Peter Xu <peterx@redhat.com>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
We traces its range, but we don't know whether it's a MAP/UNMAP. Let's
dump it as well.
Acked-by: Alex Williamson <alex.williamson@redhat.com>
Reviewed-by: David Gibson <david@gibson.dropbear.id.au>
Signed-off-by: Peter Xu <peterx@redhat.com>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
Now the kernel commit 05f0c03fbac1 ("vfio-pci: Allow to mmap
sub-page MMIO BARs if the mmio page is exclusive") allows VFIO
to mmap sub-page BARs. This is the corresponding QEMU patch.
With those patches applied, we could passthrough sub-page BARs
to guest, which can help to improve IO performance for some devices.
In this patch, we expand MemoryRegions of these sub-page
MMIO BARs to PAGE_SIZE in vfio_pci_write_config(), so that
the BARs could be passed to KVM ioctl KVM_SET_USER_MEMORY_REGION
with a valid size. The expanding size will be recovered when
the base address of sub-page BAR is changed and not page aligned
any more in guest. And we also set the priority of these BARs'
memory regions to zero in case of overlap with BARs which share
the same page with sub-page BARs in guest.
Signed-off-by: Yongji Xie <xyjxie@linux.vnet.ibm.com>
Signed-off-by: Alex Williamson <alex.williamson@redhat.com>
As reported in the link below, user has a PCI device with a 4KB BAR
which contains the MSI-X table. This seems to hit a corner case in
the kernel where the region reports being mmap capable, but the sparse
mmap information reports a zero sized range. It's not entirely clear
that the kernel is incorrect in doing this, but regardless, we need
to handle it. To do this, fill our mmap array only with non-zero
sized sparse mmap entries and add an error return from the function
so we can tell the difference between nr_mmaps being zero based on
sparse mmap info vs lack of sparse mmap info.
NB, this doesn't actually change the behavior of the device, it only
removes the scary "Failed to mmap ... Performance may be slow" error
message. We cannot currently create an mmap over the MSI-X table.
Link: http://lists.nongnu.org/archive/html/qemu-discuss/2016-10/msg00009.html
Signed-off-by: Alex Williamson <alex.williamson@redhat.com>
Setting skip_dump on a MemoryRegion allows us to modify one specific
code path, but the restriction we're trying to address encompasses
more than that. If we have a RAM MemoryRegion backed by a physical
device, it not only restricts our ability to dump that region, but
also affects how we should manipulate it. Here we recognize that
MemoryRegions do not change to sometimes allow dumps and other times
not, so we replace setting the skip_dump flag with a new initializer
so that we know exactly the type of region to which we're applying
this behavior.
Signed-off-by: Alex Williamson <alex.williamson@redhat.com>
Acked-by: Paolo Bonzini <pbonzini@redhat.com>
Pass an error object to prepare for migration to VFIO-PCI realize.
In vfio platform vfio_base_device_init we currently just report the
error. Subsequent patches will propagate the error up to the realize
function.
Signed-off-by: Eric Auger <eric.auger@redhat.com>
Reviewed-by: Markus Armbruster <armbru@redhat.com>
Signed-off-by: Alex Williamson <alex.williamson@redhat.com>
Pass an error object to prepare for migration to VFIO-PCI realize.
For the time being let's just simply report the error in
vfio platform's vfio_base_device_init(). A subsequent patch will
duly propagate the error up to vfio_platform_realize.
Signed-off-by: Eric Auger <eric.auger@redhat.com>
Reviewed-by: Markus Armbruster <armbru@redhat.com>
Signed-off-by: Alex Williamson <alex.williamson@redhat.com>
The error is currently simply reported in vfio_get_group. Don't
bother too much with the prefix which will be handled at upper level,
later on.
Also return an error value in case container->error is not 0 and
the container is teared down.
On vfio_spapr_remove_window failure, we also report an error whereas
it was silent before.
Signed-off-by: Eric Auger <eric.auger@redhat.com>
Reviewed-by: Markus Armbruster <armbru@redhat.com>
Signed-off-by: Alex Williamson <alex.williamson@redhat.com>
IOMMU Notifier list is used for notifying IO address mapping changes.
Currently VFIO is the only user.
However it is possible that future consumer like vhost would like to
only listen to part of its notifications (e.g., cache invalidations).
This patch introduced IOMMUNotifier and IOMMUNotfierFlag bits for a
finer grained control of it.
IOMMUNotifier contains a bitfield for the notify consumer describing
what kind of notification it is interested in. Currently two kinds of
notifications are defined:
- IOMMU_NOTIFIER_MAP: for newly mapped entries (additions)
- IOMMU_NOTIFIER_UNMAP: for entries to be removed (cache invalidates)
When registering the IOMMU notifier, we need to specify one or multiple
types of messages to listen to.
When notifications are triggered, its type will be checked against the
notifier's type bits, and only notifiers with registered bits will be
notified.
(For any IOMMU implementation, an in-place mapping change should be
notified with an UNMAP followed by a MAP.)
Signed-off-by: Peter Xu <peterx@redhat.com>
Message-Id: <1474606948-14391-2-git-send-email-peterx@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Tracked down with an ugly, brittle and probably buggy Perl script.
Also move includes converted to <...> up so they get included before
ours where that's obviously okay.
Signed-off-by: Markus Armbruster <armbru@redhat.com>
Reviewed-by: Eric Blake <eblake@redhat.com>
Tested-by: Eric Blake <eblake@redhat.com>
Reviewed-by: Richard Henderson <rth@twiddle.net>
New VFIO_SPAPR_TCE_v2_IOMMU type supports dynamic DMA window management.
This adds ability to VFIO common code to dynamically allocate/remove
DMA windows in the host kernel when new VFIO container is added/removed.
This adds a helper to vfio_listener_region_add which makes
VFIO_IOMMU_SPAPR_TCE_CREATE ioctl and adds just created IOMMU into
the host IOMMU list; the opposite action is taken in
vfio_listener_region_del.
When creating a new window, this uses heuristic to decide on the TCE table
levels number.
This should cause no guest visible change in behavior.
Signed-off-by: Alexey Kardashevskiy <aik@ozlabs.ru>
Reviewed-by: David Gibson <david@gibson.dropbear.id.au>
[dwg: Added some casts to prevent printf() warnings on certain targets
where the kernel headers' __u64 doesn't match uint64_t or PRIx64]
Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
There are going to be multiple IOMMUs per a container. This moves
the single host IOMMU parameter set to a list of VFIOHostDMAWindow.
This should cause no behavioral change and will be used later by
the SPAPR TCE IOMMU v2 which will also add a vfio_host_win_del() helper.
Signed-off-by: Alexey Kardashevskiy <aik@ozlabs.ru>
Reviewed-by: David Gibson <david@gibson.dropbear.id.au>
Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
This makes use of the new "memory registering" feature. The idea is
to provide the userspace ability to notify the host kernel about pages
which are going to be used for DMA. Having this information, the host
kernel can pin them all once per user process, do locked pages
accounting (once) and not spent time on doing that in real time with
possible failures which cannot be handled nicely in some cases.
This adds a prereg memory listener which listens on address_space_memory
and notifies a VFIO container about memory which needs to be
pinned/unpinned. VFIO MMIO regions (i.e. "skip dump" regions) are skipped.
The feature is only enabled for SPAPR IOMMU v2. The host kernel changes
are required. Since v2 does not need/support VFIO_IOMMU_ENABLE, this does
not call it when v2 is detected and enabled.
This enforces guest RAM blocks to be host page size aligned; however
this is not new as KVM already requires memory slots to be host page
size aligned.
Signed-off-by: Alexey Kardashevskiy <aik@ozlabs.ru>
[dwg: Fix compile error on 32-bit host]
Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
The IOMMU driver may change behavior depending on whether a notifier
client is present. In the case of POWER, this represents a change in
the visibility of the IOTLB, for other drivers such as intel-iommu and
future AMD-Vi emulation, notifier support is not yet enabled and this
provides the opportunity to flag that incompatibility.
Signed-off-by: Alexey Kardashevskiy <aik@ozlabs.ru>
Reviewed-by: David Gibson <david@gibson.dropbear.id.au>
Reviewed-by: Peter Xu <peterx@redhat.com>
Tested-by: Peter Xu <peterx@redhat.com>
Acked-by: Paolo Bonzini <pbonzini@redhat.com>
[new log & extracted from [PATCH qemu v17 12/12] spapr_iommu, vfio, memory: Notify IOMMU about starting/stopping listening]
Signed-off-by: Alex Williamson <alex.williamson@redhat.com>
Every IOMMU has some granularity which MemoryRegionIOMMUOps::translate
uses when translating, however this information is not available outside
the translate context for various checks.
This adds a get_min_page_size callback to MemoryRegionIOMMUOps and
a wrapper for it so IOMMU users (such as VFIO) can know the minimum
actual page size supported by an IOMMU.
As IOMMU MR represents a guest IOMMU, this uses TARGET_PAGE_SIZE
as fallback.
This removes vfio_container_granularity() and uses new helper in
memory_region_iommu_replay() when replaying IOMMU mappings on added
IOMMU memory region.
Signed-off-by: Alexey Kardashevskiy <aik@ozlabs.ru>
Reviewed-by: David Gibson <david@gibson.dropbear.id.au>
Acked-by: Alex Williamson <alex.williamson@redhat.com>
[dwg: Removed an unnecessary calculation]
Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
vfio_eeh_container_op() is the backend that communicates with
host kernel to support EEH functionality in QEMU. However, the
functon should return the value from host kernel instead of 0
unconditionally.
dwg: Specifically the problem occurs for the handful of EEH
sub-operations which can return a non-zero, non-error result.
Signed-off-by: Gavin Shan <gwshan@linux.vnet.ibm.com>
Acked-by: Alex Williamson <alex.williamson@redhat.com>
[dwg: clarification to commit message]
Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
qemu/osdep.h checks whether MAP_ANONYMOUS is defined, but this check
is bogus without a previous inclusion of sys/mman.h. Include it in
sysemu/os-posix.h and remove it from everywhere else.
Reviewed-by: Peter Maydell <peter.maydell@linaro.org>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
At the moment IOMMU MR only translate to the system memory.
However if some new code changes this, we will need clear indication why
it is not working so here is the check.
Signed-off-by: Alexey Kardashevskiy <aik@ozlabs.ru>
Reviewed-by: David Gibson <david@gibson.dropbear.id.au>
Signed-off-by: Alex Williamson <alex.williamson@redhat.com>
Since a788f227 "memory: Allow replay of IOMMU mapping notifications"
when new VFIO listener is added, all existing IOMMU mappings are
replayed. However there is a problem that the base address of
an IOMMU memory region (IOMMU MR) is ignored which is not a problem
for the existing user (which is pseries) with its default 32bit DMA
window starting at 0 but it is if there is another DMA window.
This stores the IOMMU's offset_within_address_space and adjusts
the IOVA before calling vfio_dma_map/vfio_dma_unmap.
As the IOMMU notifier expects IOVA offset rather than the absolute
address, this also adjusts IOVA in sPAPR H_PUT_TCE handler before
calling notifier(s).
Signed-off-by: Alexey Kardashevskiy <aik@ozlabs.ru>
Reviewed-by: David Gibson <david@gibson.dropbear.id.au>
Signed-off-by: Alex Williamson <alex.williamson@redhat.com>
7532d3cbf "vfio: Fix 128 bit handling" added support for 64bit IOMMU
memory regions when those are added to VFIO address space; however
removing code cannot cope with these as int128_get64() will fail on
1<<64.
This copies 128bit handling from region_add() to region_del().
Since the only machine type which is actually going to use 64bit IOMMU
is pseries and it never really removes them (instead it will dynamically
add/remove subregions), this should cause no behavioral change.
Signed-off-by: Alexey Kardashevskiy <aik@ozlabs.ru>
Signed-off-by: Alex Williamson <alex.williamson@redhat.com>
Given a device specific region type and sub-type, find it. Also
cleanup return point on error in vfio_get_region_info() so that we
always return 0 with a valid pointer or -errno and NULL.
Signed-off-by: Alex Williamson <alex.williamson@redhat.com>
Reviewed-by: Gerd Hoffmann <kraxel@redhat.com>
Tested-by: Gerd Hoffmann <kraxel@redhat.com>
The sparse mmap capability in a vfio region info allows vfio to tell
us which sub-areas of a region may be mmap'd. Thus rather than
assuming a single mmap covers the entire region and later frobbing it
ourselves for things like the PCI MSI-X vector table, we can read that
directly from vfio.
Signed-off-by: Alex Williamson <alex.williamson@redhat.com>
Reviewed-by: Gerd Hoffmann <kraxel@redhat.com>
Tested-by: Gerd Hoffmann <kraxel@redhat.com>
vfio_listener_region_add for a iommu mr results in
an overflow assert since iommu memory region is initialized
with UINT64_MAX. Convert calculations to 128 bit arithmetic
for iommu memory regions and let int128_get64 assert for non iommu
regions if there's an overflow.
Suggested-by: Alex Williamson <alex.williamson@redhat.com>
Signed-off-by: Bandan Das <bsd@redhat.com>
[missed (end - 1) on 2nd trace call, move llsize closer to use]
Signed-off-by: Alex Williamson <alex.williamson@redhat.com>
vfio_container_ioctl() was a bad interface that bypassed abstraction
boundaries, had semantics that sat uneasily with its name, and was unsafe
in many realistic circumstances. Now that spapr-pci-vfio-host-bridge has
been folded into spapr-pci-host-bridge, there are no more users, so remove
it.
Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
Reviewed-by: Alexey Kardashevskiy <aik@ozlabs.ru>
Acked-by: Alex Williamson <alex.williamson@redhat.com>
At present the code handling IBM's Enhanced Error Handling (EEH) interface
on VFIO devices operates by bypassing the usual VFIO logic with
vfio_container_ioctl(). That's a poorly designed interface with unclear
semantics about exactly what can be operated on.
In particular it operates on a single vfio container internally (hence the
name), but takes an address space and group id, from which it deduces the
container in a rather roundabout way. groupids are something that code
outside vfio shouldn't even be aware of.
This patch creates new interfaces for EEH operations. Internally we
have vfio_eeh_container_op() which takes a VFIOContainer object
directly. For external use we have vfio_eeh_as_ok() which determines
if an AddressSpace is usable for EEH (at present this means it has a
single container with exactly one group attached), and vfio_eeh_as_op()
which will perform an operation on an AddressSpace in the unambiguous case,
and otherwise returns an error.
This interface still isn't great, but it's enough of an improvement to
allow a number of cleanups in other places.
Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
Reviewed-by: Alexey Kardashevskiy <aik@ozlabs.ru>
Acked-by: Alex Williamson <alex.williamson@redhat.com>
Both platform and PCI vfio drivers create a "slow", I/O memory region
with one or more mmap memory regions overlayed when supported by the
device. Generalize this to a set of common helpers in the core that
pulls the region info from vfio, fills the region data, configures
slow mapping, and adds helpers for comleting the mmap, enable/disable,
and teardown. This can be immediately used by the PCI MSI-X code,
which needs to mmap around the MSI-X vector table.
This also changes VFIORegion.mem to be dynamically allocated because
otherwise we don't know how the caller has allocated VFIORegion and
therefore don't know whether to unreference it to destroy the
MemoryRegion or not.
Signed-off-by: Alex Williamson <alex.williamson@redhat.com>
In preparation for supporting capability chains on regions, wrap
ioctl(VFIO_DEVICE_GET_REGION_INFO) so we don't duplicate the code for
each caller.
Signed-off-by: Alex Williamson <alex.williamson@redhat.com>
Clean up includes so that osdep.h is included first and headers
which it implies are not included manually.
This commit was created with scripts/clean-includes.
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
Message-id: 1453832250-766-22-git-send-email-peter.maydell@linaro.org
At present the memory listener used by vfio to keep host IOMMU mappings
in sync with the guest memory image assumes that if a guest IOMMU
appears, then it has no existing mappings.
This may not be true if a VFIO device is hotplugged onto a guest bus
which didn't previously include a VFIO device, and which has existing
guest IOMMU mappings.
Therefore, use the memory_region_register_iommu_notifier_replay()
function in order to fix this case, replaying existing guest IOMMU
mappings, bringing the host IOMMU into sync with the guest IOMMU.
Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
Signed-off-by: Alex Williamson <alex.williamson@redhat.com>
Depending on the host IOMMU type we determine and record the available page
sizes for IOMMU translation. We'll need this for other validation in
future patches.
Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
Reviewed-by: Thomas Huth <thuth@redhat.com>
Reviewed-by: Laurent Vivier <lvivier@redhat.com>
Signed-off-by: Alex Williamson <alex.williamson@redhat.com>
The current vfio core code assumes that the host IOMMU is capable of
mapping any IOVA the guest wants to use to where we need. However, real
IOMMUs generally only support translating a certain range of IOVAs (the
"DMA window") not a full 64-bit address space.
The common x86 IOMMUs support a wide enough range that guests are very
unlikely to go beyond it in practice, however the IOMMU used on IBM Power
machines - in the default configuration - supports only a much more limited
IOVA range, usually 0..2GiB.
If the guest attempts to set up an IOVA range that the host IOMMU can't
map, qemu won't report an error until it actually attempts to map a bad
IOVA. If guest RAM is being mapped directly into the IOMMU (i.e. no guest
visible IOMMU) then this will show up very quickly. If there is a guest
visible IOMMU, however, the problem might not show up until much later when
the guest actually attempt to DMA with an IOVA the host can't handle.
This patch adds a test so that we will detect earlier if the guest is
attempting to use IOVA ranges that the host IOMMU won't be able to deal
with.
For now, we assume that "Type1" (x86) IOMMUs can support any IOVA, this is
incorrect, but no worse than what we have already. We can't do better for
now because the Type1 kernel interface doesn't tell us what IOVA range the
IOMMU actually supports.
For the Power "sPAPR TCE" IOMMU, however, we can retrieve the supported
IOVA range and validate guest IOVA ranges against it, and this patch does
so.
Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
Reviewed-by: Laurent Vivier <lvivier@redhat.com>
Signed-off-by: Alex Williamson <alex.williamson@redhat.com>
If a DMA mapping operation fails in vfio_listener_region_add() it
checks to see if we've already completed initial setup of the
container. If so it reports an error so the setup code can fail
gracefully, otherwise throws a hw_error().
There are other potential failure cases in vfio_listener_region_add()
which could benefit from the same logic, so move it to its own
fail: block. Later patches can use this to extend other failure cases
to fail as gracefully as possible under the circumstances.
Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
Reviewed-by: Thomas Huth <thuth@redhat.com>
Reviewed-by: Laurent Vivier <lvivier@redhat.com>
Signed-off-by: Alex Williamson <alex.williamson@redhat.com>
Currently the VFIOContainer iommu_data field contains a union with
different information for different host iommu types. However:
* It only actually contains information for the x86-like "Type1" iommu
* Because we have a common listener the Type1 fields are actually used
on all IOMMU types, including the SPAPR TCE type as well
In fact we now have a general structure for the listener which is unlikely
to ever need per-iommu-type information, so this patch removes the union.
In a similar way we can unify the setup of the vfio memory listener in
vfio_connect_container() that is currently split across a switch on iommu
type, but is effectively the same in both cases.
The iommu_data.release pointer was only needed as a cleanup function
which would handle potentially different data in the union. With the
union gone, it too can be removed.
Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
Reviewed-by: Laurent Vivier <lvivier@redhat.com>
Signed-off-by: Alex Williamson <alex.williamson@redhat.com>
The default should be to allow mmap and new drivers shouldn't need to
expose an option or set it to other than the allocation default in
their initfn. Take advantage of the experimental flag to change this
option to the correct polarity.
Signed-off-by: Alex Williamson <alex.williamson@redhat.com>
Many source files have doubled words (eg "the the", "to to",
and so on). Most of these can simply be removed, but a couple
were actual mis-spellings (eg "to to" instead of "to do").
There was even one triple word score "to to to" :-)
Signed-off-by: Daniel P. Berrange <berrange@redhat.com>
Reviewed-by: Marc-André Lureau <marcandre.lureau@redhat.com>
Reviewed-by: Markus Armbruster <armbru@redhat.com>
Signed-off-by: Michael Tokarev <mjt@tls.msk.ru>
On systems with guest visible IOMMU, adding a new memory region onto
PCI bus calls vfio_listener_region_add() for every DMA window. This
installs a notifier for IOMMU memory regions. The notifier is supposed
to be removed vfio_listener_region_del(), however in the case of mixed
PHB (emulated + VFIO devices) when last VFIO device is unplugged and
container gets destroyed, all existing DMA windows stay alive altogether
with the notifiers which are on the linked list which head was in
the destroyed container.
This unregisters IOMMU memory region notifier when a container is
destroyed.
Signed-off-by: Alexey Kardashevskiy <aik@ozlabs.ru>
Reviewed-by: David Gibson <david@gibson.dropbear.id.au>
Signed-off-by: Alex Williamson <alex.williamson@redhat.com>
Once address_space_translate will be called outside the BQL, the returned
MemoryRegion might disappear as soon as the RCU read-side critical section
ends. Avoid this by moving the critical section to the callers.
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Message-Id: <1426684909-95030-3-git-send-email-pbonzini@redhat.com>
The patch implements sPAPRPHBClass EEH callbacks so that the EEH
RTAS requests can be routed to VFIO for further handling.
Signed-off-by: Gavin Shan <gwshan@linux.vnet.ibm.com>
Reviewed-by: David Gibson <david@gibson.dropbear.id.au>
Signed-off-by: Alexander Graf <agraf@suse.de>
Disabling MMAP support uses the slower read/write accesses but allows to
trace all MMIO accesses, which is not good for performance, but very
useful for reverse engineering PCI drivers. This option allows to
disable MMAP per device without a compile-time change.
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Signed-off-by: Alex Williamson <alex.williamson@redhat.com>
They are not used from anywhere but common.c which is where these are
defined so make them static.
Signed-off-by: Alexey Kardashevskiy <aik@ozlabs.ru>
Signed-off-by: Alex Williamson <alex.williamson@redhat.com>
This makes the error report more informative.
Signed-off-by: Alexey Kardashevskiy <aik@ozlabs.ru>
Signed-off-by: Alex Williamson <alex.williamson@redhat.com>
The difference between v1 and v2 is fairly subtle, simply more
deterministic behavior for unmaps. The v1 interface allows the user
to attempt to unmap sub-regions of previous mappings, returning
success with zero size if unable to comply. This was a reflection of
the underlying IOMMU API. The v2 interface requires that the user
may only unmap fully contained mappings, ie. an unmap cannot intersect
or bisect a previous mapping, but may cover multiple mappings. QEMU
never made use of the sub-region v1 support anyway, so we can support
either v1 or v2. We'll favor v2 since it's newer.
Signed-off-by: Alex Williamson <alex.williamson@redhat.com>
In order to enable out-of-BQL address space lookup, destruction of
devices needs to be split in two phases.
Unrealize is the first phase; once it complete no new accesses will
be started, but there may still be pending memory accesses can still
be completed.
The second part is freeing the device, which only happens once all memory
accesses are complete. At this point the reference count has dropped to
zero, an RCU grace period must have completed (because the RCU-protected
FlatViews hold a reference to the device via memory_region_ref). This is
when instance_finalize is called.
Freeing data belongs in an instance_finalize callback, because the
dynamically allocated memory can still be used after unrealize by the
pending memory accesses.
This starts the process by creating an instance_finalize callback and
freeing most of the dynamically-allocated data in instance_finalize.
Because instance_finalize is also called on error paths or also when
the device is actually not realized, the common code needs some changes
to be ready for this. The error path in vfio_initfn can be simplified too.
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Signed-off-by: Alex Williamson <alex.williamson@redhat.com>
Now that vfio_put_base_device is called unconditionally at instance_finalize
time, it can be called twice if vfio_populate_device fails. This works
but it is slightly harder to follow.
Change vfio_get_device to not touch the vbasedev struct until it will
definitely succeed, moving the vfio_populate_device call back to vfio-pci.
This way, vfio_put_base_device will only be called once.
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Signed-off-by: Alex Williamson <alex.williamson@redhat.com>
With the conversion to tracepoints, a couple previous DPRINTKs are
now quite a bit more visible and are really just informational.
Remove these and add a bit more description to another.
Signed-off-by: Alex Williamson <alex.williamson@redhat.com>
A new common module is created. It implements all functions
that have no device specificity (PCI, Platform).
This patch only consists in move (no functional changes)
Signed-off-by: Kim Phillips <kim.phillips@linaro.org>
Signed-off-by: Eric Auger <eric.auger@linaro.org>
Signed-off-by: Alex Williamson <alex.williamson@redhat.com>
