We support coordinated discarding of RAM using the RamDiscardManager for
the VFIO_TYPE1 iommus. Let's unlock support for coordinated discards,
keeping uncoordinated discards (e.g., via virtio-balloon) disabled if
possible.
This unlocks virtio-mem + vfio on x86-64. Note that vfio used via "nvme://"
by the block layer has to be implemented/unlocked separately. For now,
virtio-mem only supports x86-64; we don't restrict RamDiscardManager to
x86-64, though: arm64 and s390x are supposed to work as well, and we'll
test once unlocking virtio-mem support. The spapr IOMMUs will need special
care, to be tackled later, e.g.., once supporting virtio-mem.
Note: The block size of a virtio-mem device has to be set to sane sizes,
depending on the maximum hotplug size - to not run out of vfio mappings.
The default virtio-mem block size is usually in the range of a couple of
MBs. The maximum number of mapping is 64k, shared with other users.
Assume you want to hotplug 256GB using virtio-mem - the block size would
have to be set to at least 8 MiB (resulting in 32768 separate mappings).
Acked-by: Alex Williamson <alex.williamson@redhat.com>
Reviewed-by: Alex Williamson <alex.williamson@redhat.com>
Acked-by: Michael S. Tsirkin <mst@redhat.com>
Cc: Paolo Bonzini <pbonzini@redhat.com>
Cc: "Michael S. Tsirkin" <mst@redhat.com>
Cc: Alex Williamson <alex.williamson@redhat.com>
Cc: Dr. David Alan Gilbert <dgilbert@redhat.com>
Cc: Igor Mammedov <imammedo@redhat.com>
Cc: Pankaj Gupta <pankaj.gupta.linux@gmail.com>
Cc: Peter Xu <peterx@redhat.com>
Cc: Auger Eric <eric.auger@redhat.com>
Cc: Wei Yang <richard.weiyang@linux.alibaba.com>
Cc: teawater <teawaterz@linux.alibaba.com>
Cc: Marek Kedzierski <mkedzier@redhat.com>
Signed-off-by: David Hildenbrand <david@redhat.com>
Message-Id: <20210413095531.25603-14-david@redhat.com>
Signed-off-by: Eduardo Habkost <ehabkost@redhat.com>
vIOMMU support works already with RamDiscardManager as long as guests only
map populated memory. Both, populated and discarded memory is mapped
into &address_space_memory, where vfio_get_xlat_addr() will find that
memory, to create the vfio mapping.
Sane guests will never map discarded memory (e.g., unplugged memory
blocks in virtio-mem) into an IOMMU - or keep it mapped into an IOMMU while
memory is getting discarded. However, there are two cases where a malicious
guests could trigger pinning of more memory than intended.
One case is easy to handle: the guest trying to map discarded memory
into an IOMMU.
The other case is harder to handle: the guest keeping memory mapped in
the IOMMU while it is getting discarded. We would have to walk over all
mappings when discarding memory and identify if any mapping would be a
violation. Let's keep it simple for now and print a warning, indicating
that setting RLIMIT_MEMLOCK can mitigate such attacks.
We have to take care of incoming migration: at the point the
IOMMUs get restored and start creating mappings in vfio, RamDiscardManager
implementations might not be back up and running yet: let's add runstate
priorities to enforce the order when restoring.
Acked-by: Alex Williamson <alex.williamson@redhat.com>
Reviewed-by: Alex Williamson <alex.williamson@redhat.com>
Acked-by: Michael S. Tsirkin <mst@redhat.com>
Cc: Paolo Bonzini <pbonzini@redhat.com>
Cc: "Michael S. Tsirkin" <mst@redhat.com>
Cc: Alex Williamson <alex.williamson@redhat.com>
Cc: Dr. David Alan Gilbert <dgilbert@redhat.com>
Cc: Igor Mammedov <imammedo@redhat.com>
Cc: Pankaj Gupta <pankaj.gupta.linux@gmail.com>
Cc: Peter Xu <peterx@redhat.com>
Cc: Auger Eric <eric.auger@redhat.com>
Cc: Wei Yang <richard.weiyang@linux.alibaba.com>
Cc: teawater <teawaterz@linux.alibaba.com>
Cc: Marek Kedzierski <mkedzier@redhat.com>
Signed-off-by: David Hildenbrand <david@redhat.com>
Message-Id: <20210413095531.25603-10-david@redhat.com>
Signed-off-by: Eduardo Habkost <ehabkost@redhat.com>
Although RamDiscardManager can handle running into the maximum number of
DMA mappings by propagating errors when creating a DMA mapping, we want
to sanity check and warn the user early that there is a theoretical setup
issue and that virtio-mem might not be able to provide as much memory
towards a VM as desired.
As suggested by Alex, let's use the number of KVM memory slots to guess
how many other mappings we might see over time.
Acked-by: Alex Williamson <alex.williamson@redhat.com>
Reviewed-by: Alex Williamson <alex.williamson@redhat.com>
Acked-by: Michael S. Tsirkin <mst@redhat.com>
Cc: Paolo Bonzini <pbonzini@redhat.com>
Cc: "Michael S. Tsirkin" <mst@redhat.com>
Cc: Alex Williamson <alex.williamson@redhat.com>
Cc: Dr. David Alan Gilbert <dgilbert@redhat.com>
Cc: Igor Mammedov <imammedo@redhat.com>
Cc: Pankaj Gupta <pankaj.gupta.linux@gmail.com>
Cc: Peter Xu <peterx@redhat.com>
Cc: Auger Eric <eric.auger@redhat.com>
Cc: Wei Yang <richard.weiyang@linux.alibaba.com>
Cc: teawater <teawaterz@linux.alibaba.com>
Cc: Marek Kedzierski <mkedzier@redhat.com>
Signed-off-by: David Hildenbrand <david@redhat.com>
Message-Id: <20210413095531.25603-9-david@redhat.com>
Signed-off-by: Eduardo Habkost <ehabkost@redhat.com>
Let's query the maximum number of possible DMA mappings by querying the
available mappings when creating the container (before any mappings are
created). We'll use this informaton soon to perform some sanity checks
and warn the user.
Reviewed-by: Alex Williamson <alex.williamson@redhat.com>
Acked-by: Alex Williamson <alex.williamson@redhat.com>
Acked-by: Michael S. Tsirkin <mst@redhat.com>
Cc: Paolo Bonzini <pbonzini@redhat.com>
Cc: "Michael S. Tsirkin" <mst@redhat.com>
Cc: Alex Williamson <alex.williamson@redhat.com>
Cc: Dr. David Alan Gilbert <dgilbert@redhat.com>
Cc: Igor Mammedov <imammedo@redhat.com>
Cc: Pankaj Gupta <pankaj.gupta.linux@gmail.com>
Cc: Peter Xu <peterx@redhat.com>
Cc: Auger Eric <eric.auger@redhat.com>
Cc: Wei Yang <richard.weiyang@linux.alibaba.com>
Cc: teawater <teawaterz@linux.alibaba.com>
Cc: Marek Kedzierski <mkedzier@redhat.com>
Signed-off-by: David Hildenbrand <david@redhat.com>
Message-Id: <20210413095531.25603-8-david@redhat.com>
Signed-off-by: Eduardo Habkost <ehabkost@redhat.com>
Implement support for RamDiscardManager, to prepare for virtio-mem
support. Instead of mapping the whole memory section, we only map
"populated" parts and update the mapping when notified about
discarding/population of memory via the RamDiscardListener. Similarly, when
syncing the dirty bitmaps, sync only the actually mapped (populated) parts
by replaying via the notifier.
Using virtio-mem with vfio is still blocked via
ram_block_discard_disable()/ram_block_discard_require() after this patch.
Reviewed-by: Alex Williamson <alex.williamson@redhat.com>
Acked-by: Alex Williamson <alex.williamson@redhat.com>
Acked-by: Michael S. Tsirkin <mst@redhat.com>
Cc: Paolo Bonzini <pbonzini@redhat.com>
Cc: "Michael S. Tsirkin" <mst@redhat.com>
Cc: Alex Williamson <alex.williamson@redhat.com>
Cc: Dr. David Alan Gilbert <dgilbert@redhat.com>
Cc: Igor Mammedov <imammedo@redhat.com>
Cc: Pankaj Gupta <pankaj.gupta.linux@gmail.com>
Cc: Peter Xu <peterx@redhat.com>
Cc: Auger Eric <eric.auger@redhat.com>
Cc: Wei Yang <richard.weiyang@linux.alibaba.com>
Cc: teawater <teawaterz@linux.alibaba.com>
Cc: Marek Kedzierski <mkedzier@redhat.com>
Signed-off-by: David Hildenbrand <david@redhat.com>
Message-Id: <20210413095531.25603-7-david@redhat.com>
Signed-off-by: Eduardo Habkost <ehabkost@redhat.com>
For now the switch of vfio dirty page tracking is integrated into
@vfio_save_handler. The reason is that some PCI vendor driver may
start to track dirty base on _SAVING state of device, so if dirty
tracking is started before setting device state, vfio will report
full-dirty to QEMU.
However, the dirty bmap of all ramblocks are fully set when setup
ram saving, so it's not matter whether the device is in _SAVING
state when start vfio dirty tracking.
Moreover, this logic causes some problems [1]. The object of dirty
tracking is guest memory, but the object of @vfio_save_handler is
device state, which produces unnecessary coupling and conflicts:
1. Coupling: Their saving granule is different (perVM vs perDevice).
vfio will enable dirty_page_tracking for each devices, actually
once is enough.
2. Conflicts: The ram_save_setup() traverses all memory_listeners
to execute their log_start() and log_sync() hooks to get the
first round dirty bitmap, which is used by the bulk stage of
ram saving. However, as vfio dirty tracking is not yet started,
it can't get dirty bitmap from vfio. Then we give up the chance
to handle vfio dirty page at bulk stage.
Move the switch of vfio dirty_page_tracking into vfio_memory_listener
can solve above problems. Besides, Do not require devices in SAVING
state for vfio_sync_dirty_bitmap().
[1] https://www.spinics.net/lists/kvm/msg229967.html
Reported-by: Zenghui Yu <yuzenghui@huawei.com>
Signed-off-by: Keqian Zhu <zhukeqian1@huawei.com>
Suggested-by: Paolo Bonzini <pbonzini@redhat.com>
Message-Id: <20210309031913.11508-1-zhukeqian1@huawei.com>
Signed-off-by: Alex Williamson <alex.williamson@redhat.com>
The cpu_physical_memory_set_dirty_lebitmap() can quickly deal with
the dirty pages of memory by bitmap-traveling, regardless of whether
the bitmap is aligned correctly or not.
cpu_physical_memory_set_dirty_lebitmap() supports pages in bitmap of
host page size. So it'd better to set bitmap_pgsize to host page size
to support more translation granule sizes.
[aw: The Fixes commit below introduced code to restrict migration
support to configurations where the target page size intersects the
host dirty page support. For example, a 4K guest on a 4K host.
Due to the above flexibility in bitmap handling, this restriction
unnecessarily prevents mixed target/host pages size that could
otherwise be supported. Use host page size for dirty bitmap.]
Fixes: 87ea529c50 ("vfio: Get migration capability flags for container")
Signed-off-by: Kunkun Jiang <jiangkunkun@huawei.com>
Message-Id: <20210304133446.1521-1-jiangkunkun@huawei.com>
Signed-off-by: Alex Williamson <alex.williamson@redhat.com>
In an attempt to fix smmu/virtio-iommu - vhost regression, commit
958ec334bc ("vhost: Unbreak SMMU and virtio-iommu on dev-iotlb support")
broke virtio-iommu integration. This is due to the fact VFIO registers
IOMMU_NOTIFIER_ALL notifiers, which includes IOMMU_NOTIFIER_DEVIOTLB_UNMAP
and this latter now is rejected by the virtio-iommu. As a consequence,
the registration fails. VHOST behaves like a device with an ATC cache. The
VFIO device does not support this scheme yet.
Let's register only legacy MAP and UNMAP notifiers.
Fixes: 958ec334bc ("vhost: Unbreak SMMU and virtio-iommu on dev-iotlb support")
Signed-off-by: Eric Auger <eric.auger@redhat.com>
Message-Id: <20210209213233.40985-2-eric.auger@redhat.com>
Acked-by: Jason Wang <jasowang@redhat.com>
Acked-by: Alex Williamson <alex.williamson@redhat.com>
Signed-off-by: Alex Williamson <alex.williamson@redhat.com>
There is an obvious typo in the function name of the .log_sync() callback.
Spell it correctly.
Signed-off-by: Zenghui Yu <yuzenghui@huawei.com>
Message-Id: <20201204014240.772-1-yuzenghui@huawei.com>
Signed-off-by: Alex Williamson <alex.williamson@redhat.com>
By default dirty pages tracking is enabled during iterative phase
(pre-copy phase).
Added per device opt-out option 'x-pre-copy-dirty-page-tracking' to
disable dirty pages tracking during iterative phase. If the option
'x-pre-copy-dirty-page-tracking=off' is set for any VFIO device, dirty
pages tracking during iterative phase will be disabled.
Signed-off-by: Kirti Wankhede <kwankhede@nvidia.com>
Signed-off-by: Alex Williamson <alex.williamson@redhat.com>
IOMMUs may declare memory regions spanning from 0 to UINT64_MAX. When
attempting to deal with such region, vfio_listener_region_del() passes a
size of 2^64 to int128_get64() which throws an assertion failure. Even
ignoring this, the VFIO_IOMMU_DMA_MAP ioctl cannot handle this size
since the size field is 64-bit. Split the request in two.
Acked-by: Alex Williamson <alex.williamson@redhat.com>
Reviewed-by: Eric Auger <eric.auger@redhat.com>
Signed-off-by: Jean-Philippe Brucker <jean-philippe@linaro.org>
Message-Id: <20201030180510.747225-11-jean-philippe@linaro.org>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
Set IOMMU supported page size mask same as host Linux supported page
size mask.
Acked-by: Alex Williamson <alex.williamson@redhat.com>
Reviewed-by: Eric Auger <eric.auger@redhat.com>
Signed-off-by: Bharat Bhushan <bbhushan2@marvell.com>
Signed-off-by: Jean-Philippe Brucker <jean-philippe@linaro.org>
Message-Id: <20201030180510.747225-9-jean-philippe@linaro.org>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
The type of input variable is unsigned int
while the printer type is int. So fix incorrect print type.
Signed-off-by: Zhengui li <lizhengui@huawei.com>
Signed-off-by: Alex Williamson <alex.williamson@redhat.com>
Now that VFIO_DEVICE_GET_INFO supports capability chains, add a helper
function to find specific capabilities in the chain.
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>
The underlying host may be limiting the number of outstanding DMA
requests for type 1 IOMMU. Add helper functions to check for the
DMA available capability and retrieve the current number of DMA
mappings allowed.
Signed-off-by: Matthew Rosato <mjrosato@linux.ibm.com>
Reviewed-by: Cornelia Huck <cohuck@redhat.com>
[aw: vfio_get_info_dma_avail moved inside CONFIG_LINUX]
Signed-off-by: Alex Williamson <alex.williamson@redhat.com>
Rather than duplicating the same loop in multiple locations,
create a static function to do the work.
Signed-off-by: Matthew Rosato <mjrosato@linux.ibm.com>
Reviewed-by: Philippe Mathieu-Daudé <philmd@redhat.com>
Reviewed-by: Cornelia Huck <cohuck@redhat.com>
Signed-off-by: Alex Williamson <alex.williamson@redhat.com>
Added amount of bytes transferred to the VM at destination by all VFIO
devices
Signed-off-by: Kirti Wankhede <kwankhede@nvidia.com>
Reviewed-by: Dr. David Alan Gilbert <dgilbert@redhat.com>
Signed-off-by: Alex Williamson <alex.williamson@redhat.com>
With vIOMMU, IO virtual address range can get unmapped while in pre-copy
phase of migration. In that case, unmap ioctl should return pages pinned
in that range and QEMU should find its correcponding guest physical
addresses and report those dirty.
Suggested-by: Alex Williamson <alex.williamson@redhat.com>
Signed-off-by: Kirti Wankhede <kwankhede@nvidia.com>
Reviewed-by: Neo Jia <cjia@nvidia.com>
[aw: fix error_report types, fix cpu_physical_memory_set_dirty_lebitmap() cast]
Signed-off-by: Alex Williamson <alex.williamson@redhat.com>
When vIOMMU is enabled, register MAP notifier from log_sync when all
devices in container are in stop and copy phase of migration. Call replay
and get dirty pages from notifier callback.
Suggested-by: Alex Williamson <alex.williamson@redhat.com>
Signed-off-by: Kirti Wankhede <kwankhede@nvidia.com>
Reviewed-by: Yan Zhao <yan.y.zhao@intel.com>
Signed-off-by: Alex Williamson <alex.williamson@redhat.com>
vfio_listener_log_sync gets list of dirty pages from container using
VFIO_IOMMU_GET_DIRTY_BITMAP ioctl and mark those pages dirty when all
devices are stopped and saving state.
Return early for the RAM block section of mapped MMIO region.
Signed-off-by: Kirti Wankhede <kwankhede@nvidia.com>
Reviewed-by: Neo Jia <cjia@nvidia.com>
[aw: fix error_report types, fix cpu_physical_memory_set_dirty_lebitmap() cast]
Signed-off-by: Alex Williamson <alex.williamson@redhat.com>
Added helper functions to get IOMMU info capability chain.
Added function to get migration capability information from that
capability chain for IOMMU container.
Similar change was proposed earlier:
https://lists.gnu.org/archive/html/qemu-devel/2018-05/msg03759.html
Disable migration for devices if IOMMU module doesn't support migration
capability.
Signed-off-by: Kirti Wankhede <kwankhede@nvidia.com>
Cc: Shameer Kolothum <shameerali.kolothum.thodi@huawei.com>
Cc: Eric Auger <eric.auger@redhat.com>
Signed-off-by: Alex Williamson <alex.williamson@redhat.com>
This function will be used for migration region.
Migration region is mmaped when migration starts and will be unmapped when
migration is complete.
Signed-off-by: Kirti Wankhede <kwankhede@nvidia.com>
Reviewed-by: Neo Jia <cjia@nvidia.com>
Reviewed-by: Cornelia Huck <cohuck@redhat.com>
Signed-off-by: Alex Williamson <alex.williamson@redhat.com>
We want to introduce a new version of qemu_open() that uses an Error
object for reporting problems and make this it the preferred interface.
Rename the existing method to release the namespace for the new impl.
Reviewed-by: Eric Blake <eblake@redhat.com>
Reviewed-by: Philippe Mathieu-Daudé <philmd@redhat.com>
Reviewed-by: Markus Armbruster <armbru@redhat.com>
Signed-off-by: Daniel P. Berrangé <berrange@redhat.com>
VFIO is (except devices without a physical IOMMU or some mediated devices)
incompatible with discarding of RAM. The kernel will pin basically all VM
memory. Let's convert to ram_block_discard_disable(), which can now
fail, in contrast to qemu_balloon_inhibit().
Leave "x-balloon-allowed" named as it is for now.
Reviewed-by: Tony Krowiak <akrowiak@linux.ibm.com>
Acked-by: Cornelia Huck <cohuck@redhat.com>
Cc: Cornelia Huck <cohuck@redhat.com>
Cc: Alex Williamson <alex.williamson@redhat.com>
Cc: Christian Borntraeger <borntraeger@de.ibm.com>
Cc: Tony Krowiak <akrowiak@linux.ibm.com>
Cc: Halil Pasic <pasic@linux.ibm.com>
Cc: Pierre Morel <pmorel@linux.ibm.com>
Cc: Eric Farman <farman@linux.ibm.com>
Signed-off-by: David Hildenbrand <david@redhat.com>
Message-Id: <20200626072248.78761-4-david@redhat.com>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
In a few places we report errno formatted as a negative integer.
This is not as user friendly as it can be. Use strerror() and/or
error_setg_errno() instead.
Signed-off-by: Michal Privoznik <mprivozn@redhat.com>
Reviewed-by: Ján Tomko <jtomko@redhat.com>
Reviewed-by: Cornelia Huck <cohuck@redhat.com>
Reviewed-by: Eric Auger <eric.auger@redhat.com>
Reviewed-by: Philippe Mathieu-Daudé <philmd@redhat.com>
Reviewed-by: Alex Williamson <alex.williamson@redhat.com>
Message-Id: <4949c3ecf1a32189b8a4b5eb4b0fd04c1122501d.1581674006.git.mprivozn@redhat.com>
Signed-off-by: Laurent Vivier <laurent@vivier.eu>
Currently, when a notifier is attempted to be registered and its
flags are not supported (especially the MAP one) by the IOMMU MR,
we generally abruptly exit in the IOMMU code. The failure could be
handled more nicely in the caller and especially in the VFIO code.
So let's allow memory_region_register_iommu_notifier() to fail as
well as notify_flag_changed() callback.
All sites implementing the callback are updated. This patch does
not yet remove the exit(1) in the amd_iommu code.
in SMMUv3 we turn the warning message into an error message saying
that the assigned device would not work properly.
Signed-off-by: Eric Auger <eric.auger@redhat.com>
Reviewed-by: Peter Xu <peterx@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
The container error integer field is currently used to store
the first error potentially encountered during any
vfio_listener_region_add() call. However this fails to propagate
detailed error messages up to the vfio_connect_container caller.
Instead of using an integer, let's use an Error handle.
Messages are slightly reworded to accomodate the propagation.
Signed-off-by: Eric Auger <eric.auger@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
In my "build everything" tree, changing qemu/main-loop.h triggers a
recompile of some 5600 out of 6600 objects (not counting tests and
objects that don't depend on qemu/osdep.h). It includes block/aio.h,
which in turn includes qemu/event_notifier.h, qemu/notify.h,
qemu/processor.h, qemu/qsp.h, qemu/queue.h, qemu/thread-posix.h,
qemu/thread.h, qemu/timer.h, and a few more.
Include qemu/main-loop.h only where it's needed. Touching it now
recompiles only some 1700 objects. For block/aio.h and
qemu/event_notifier.h, these numbers drop from 5600 to 2800. For the
others, they shrink only slightly.
Signed-off-by: Markus Armbruster <armbru@redhat.com>
Message-Id: <20190812052359.30071-21-armbru@redhat.com>
Reviewed-by: Alex Bennée <alex.bennee@linaro.org>
Reviewed-by: Philippe Mathieu-Daudé <philmd@redhat.com>
Tested-by: Philippe Mathieu-Daudé <philmd@redhat.com>
In my "build everything" tree, changing sysemu/reset.h triggers a
recompile of some 2600 out of 6600 objects (not counting tests and
objects that don't depend on qemu/osdep.h).
The main culprit is hw/hw.h, which supposedly includes it for
convenience.
Include sysemu/reset.h only where it's needed. Touching it now
recompiles less than 200 objects.
Signed-off-by: Markus Armbruster <armbru@redhat.com>
Reviewed-by: Philippe Mathieu-Daudé <philmd@redhat.com>
Reviewed-by: Alistair Francis <alistair.francis@wdc.com>
Tested-by: Philippe Mathieu-Daudé <philmd@redhat.com>
Message-Id: <20190812052359.30071-9-armbru@redhat.com>
The code used to assign an interrupt index/subindex to an
eventfd is duplicated many times. Let's introduce an helper that
allows to set/unset the signaling for an ACTION_TRIGGER,
ACTION_MASK or ACTION_UNMASK action.
In the error message, we now use errno in case of any
VFIO_DEVICE_SET_IRQS ioctl failure.
Signed-off-by: Eric Auger <eric.auger@redhat.com>
Reviewed-by: Cornelia Huck <cohuck@redhat.com>
Reviewed-by: Li Qiang <liq3ea@gmail.com>
Signed-off-by: Alex Williamson <alex.williamson@redhat.com>
This makes vfio_get_region_info_cap() to be used in quirks.
Signed-off-by: Alexey Kardashevskiy <aik@ozlabs.ru>
Acked-by: Alex Williamson <alex.williamson@redhat.com>
Message-Id: <20190307050518.64968-3-aik@ozlabs.ru>
Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
We introduce the vfio_init_container_type() helper.
It computes the highest usable iommu type and then
set the container and the iommu type.
Its usage in vfio_connect_container() makes the code
ready for addition of new iommu types.
Signed-off-by: Eric Auger <eric.auger@redhat.com>
Reviewed-by: Greg Kurz <groug@kaod.org>
Signed-off-by: Alex Williamson <alex.williamson@redhat.com>
A kernel bug was introduced in v4.15 via commit 71a7d3d78e3c which
adds a test for address space wrap-around in the vfio DMA unmap path.
Unfortunately due to overflow, the kernel detects an unmap of the last
page in the 64-bit address space as a wrap-around. In QEMU, a Q35
guest with VT-d emulation and guest IOMMU enabled will attempt to make
such an unmap request during VM system reset, triggering an error:
qemu-kvm: VFIO_UNMAP_DMA: -22
qemu-kvm: vfio_dma_unmap(0x561f059948f0, 0xfef00000, 0xffffffff01100000) = -22 (Invalid argument)
Here the IOVA start address (0xfef00000) and the size parameter
(0xffffffff01100000) add to exactly 2^64, triggering the bug. A
kernel fix is queued for the Linux v5.0 release to address this.
This patch implements a workaround to retry the unmap, excluding the
final page of the range when we detect an unmap failing which matches
the requirements for this issue. This is expected to be a safe and
complete workaround as the VT-d address space does not extend to the
full 64-bit space and therefore the last page should never be mapped.
This workaround can be removed once all kernels with this bug are
sufficiently deprecated.
Link: https://bugzilla.redhat.com/show_bug.cgi?id=1662291
Reported-by: Pei Zhang <pezhang@redhat.com>
Debugged-by: Peter Xu <peterx@redhat.com>
Reviewed-by: Peter Xu <peterx@redhat.com>
Reviewed-by: Cornelia Huck <cohuck@redhat.com>
Signed-off-by: Alex Williamson <alex.williamson@redhat.com>
This will be needed when we change the QTAILQ head and elem structs
to unions. However, it is also consistent with the usage elsewhere
in QEMU for other list head structs (see for example FsMountList).
Note that most QTAILQs only need their name in order to do backwards
walks. Those do not break with the struct->union change, and anyway
the change will also remove the need to name heads when doing backwards
walks, so those are not touched here.
Reviewed-by: Philippe Mathieu-Daudé <philmd@redhat.com>
Reviewed-by: Markus Armbruster <armbru@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
It is not used outside hw/vfio/common.c, so it does not need to
be extern.
Acked-by: Alex Williamson <alex.williamson@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
A new error path fails to close the device file descriptor when
triggered by a ballooning incompatibility within the group. Fix it.
Fixes: 238e917285 ("vfio/ccw/pci: Allow devices to opt-in for ballooning")
Reviewed-by: Peter Xu <peterx@redhat.com>
Signed-off-by: Alex Williamson <alex.williamson@redhat.com>
At the moment the PPC64/pseries guest only supports 4K/64K/16M IOMMU
pages and POWER8 CPU supports the exact same set of page size so
so far things worked fine.
However POWER9 supports different set of sizes - 4K/64K/2M/1G and
the last two - 2M and 1G - are not even allowed in the paravirt interface
(RTAS DDW) so we always end up using 64K IOMMU pages, although we could
back guest's 16MB IOMMU pages with 2MB pages on the host.
This stores the supported host IOMMU page sizes in VFIOContainer and uses
this later when creating a new DMA window. This uses the system page size
(64k normally, 2M/16M/1G if hugepages used) as the upper limit of
the IOMMU pagesize.
This changes the type of @pagesize to uint64_t as this is what
memory_region_iommu_get_min_page_size() returns and clz64() takes.
There should be no behavioral changes on platforms other than pseries.
The guest will keep using the IOMMU page size selected by the PHB pagesize
property as this only changes the underlying hardware TCE table
granularity.
Signed-off-by: Alexey Kardashevskiy <aik@ozlabs.ru>
Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
If a vfio assigned device makes use of a physical IOMMU, then memory
ballooning is necessarily inhibited due to the page pinning, lack of
page level granularity at the IOMMU, and sufficient notifiers to both
remove the page on balloon inflation and add it back on deflation.
However, not all devices are backed by a physical IOMMU. In the case
of mediated devices, if a vendor driver is well synchronized with the
guest driver, such that only pages actively used by the guest driver
are pinned by the host mdev vendor driver, then there should be no
overlap between pages available for the balloon driver and pages
actively in use by the device. Under these conditions, ballooning
should be safe.
vfio-ccw devices are always mediated devices and always operate under
the constraints above. Therefore we can consider all vfio-ccw devices
as balloon compatible.
The situation is far from straightforward with vfio-pci. These
devices can be physical devices with physical IOMMU backing or
mediated devices where it is unknown whether a physical IOMMU is in
use or whether the vendor driver is well synchronized to the working
set of the guest driver. The safest approach is therefore to assume
all vfio-pci devices are incompatible with ballooning, but allow user
opt-in should they have further insight into mediated devices.
Signed-off-by: Alex Williamson <alex.williamson@redhat.com>
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>