In order to reduce excessive memory mapping cost in live migration and
VM reboot, it is desirable to decouple the vhost-vdpa IOTLB abstraction
from the virtio device life cycle, i.e. mappings can be kept intact
across virtio device reset. Leverage the .reset_map callback, which is
meant to destroy the iotlb on the given ASID and recreate the 1:1
passthrough/identity mapping. To be consistent, the mapping on device
creation is initiailized to passthrough/identity with PA 1:1 mapped as
IOVA. With this the device .reset op doesn't have to maintain and clean
up memory mappings by itself.
Additionally, implement .compat_reset to cater for older userspace,
which may wish to see mapping to be cleared during reset.
Signed-off-by: Si-Wei Liu <si-wei.liu@oracle.com>
Tested-by: Stefano Garzarella <sgarzare@redhat.com>
Message-Id: <1697880319-4937-8-git-send-email-si-wei.liu@oracle.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
Tested-by: Lei Yang <leiyang@redhat.com>
Since commit 6f5312f801 ("vdpa/mlx5: Add support for running with
virtio_vdpa"), mlx5_vdpa starts with preallocate 1:1 DMA MR at device
creation time. This 1:1 DMA MR will be implicitly destroyed while the
first .set_map call is invoked, in which case callers like vhost-vdpa
will start to set up custom mappings. When the .reset callback is
invoked, the custom mappings will be cleared and the 1:1 DMA MR will be
re-created.
In order to reduce excessive memory mapping cost in live migration, it
is desirable to decouple the vhost-vdpa IOTLB abstraction from the
virtio device life cycle, i.e. mappings can be kept around intact across
virtio device reset. Leverage the .reset_map callback, which is meant to
destroy the regular MR (including cvq mapping) on the given ASID and
recreate the initial DMA mapping. That way, the device .reset op runs
free from having to maintain and clean up memory mappings by itself.
Additionally, implement .compat_reset to cater for older userspace,
which may wish to see mapping to be cleared during reset.
Co-developed-by: Dragos Tatulea <dtatulea@nvidia.com>
Signed-off-by: Dragos Tatulea <dtatulea@nvidia.com>
Signed-off-by: Si-Wei Liu <si-wei.liu@oracle.com>
Message-Id: <1697880319-4937-7-git-send-email-si-wei.liu@oracle.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
Tested-by: Lei Yang <leiyang@redhat.com>
Offer this backend feature as mlx5 is compatible with it. It allows it
to do live migration with CVQ, dynamically switching between passthrough
and shadow virtqueue.
Signed-off-by: Eugenio Pérez <eperezma@redhat.com>
Message-Id: <20230703142514.363256-1-eperezma@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
For the following sequence:
- cvq group is in ASID 0
- .set_map(1, cvq_iotlb)
- .set_group_asid(cvq_group, 1)
... the cvq mapping from ASID 0 will be used. This is not always correct
behaviour.
This patch adds support for the above mentioned flow by saving the iotlb
on each .set_map and updating the cvq iotlb with it on a cvq group change.
Acked-by: Jason Wang <jasowang@redhat.com>
Acked-by: Eugenio Pérez <eperezma@redhat.com>
Signed-off-by: Dragos Tatulea <dtatulea@nvidia.com>
Message-Id: <20231018171456.1624030-18-dtatulea@nvidia.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
Reviewed-by: Si-Wei Liu <si-wei.liu@oracle.com>
Tested-by: Si-Wei Liu <si-wei.liu@oracle.com>
Tested-by: Lei Yang <leiyang@redhat.com>
Vq descriptor mappings are supported in hardware by filling in an
additional mkey which contains the descriptor mappings to the hw vq.
A previous patch in this series added support for hw mkey (mr) creation
for ASID 1.
This patch fills in both the vq data and vq descriptor mkeys based on
group ASID mapping.
The feature is signaled to the vdpa core through the presence of the
.get_vq_desc_group op.
Acked-by: Jason Wang <jasowang@redhat.com>
Acked-by: Eugenio Pérez <eperezma@redhat.com>
Signed-off-by: Dragos Tatulea <dtatulea@nvidia.com>
Message-Id: <20231018171456.1624030-16-dtatulea@nvidia.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
Reviewed-by: Si-Wei Liu <si-wei.liu@oracle.com>
Tested-by: Si-Wei Liu <si-wei.liu@oracle.com>
Tested-by: Lei Yang <leiyang@redhat.com>
The current flow for updating an mr works directly on mvdev->mr which
makes it cumbersome to handle multiple new mr structs.
This patch makes the flow more straightforward by having
mlx5_vdpa_create_mr return a new mr which will update the old mr (if
any). The old mr will be deleted and unlinked from mvdev. For the case
when the iotlb is empty (not NULL), the old mr will be cleared.
This change paves the way for adding mrs for different ASIDs.
The initialized bool is no longer needed as mr is now a pointer in the
mlx5_vdpa_dev struct which will be NULL when not initialized.
Acked-by: Eugenio Pérez <eperezma@redhat.com>
Acked-by: Jason Wang <jasowang@redhat.com>
Signed-off-by: Dragos Tatulea <dtatulea@nvidia.com>
Message-Id: <20231018171456.1624030-14-dtatulea@nvidia.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
Reviewed-by: Si-Wei Liu <si-wei.liu@oracle.com>
Tested-by: Si-Wei Liu <si-wei.liu@oracle.com>
Tested-by: Lei Yang <leiyang@redhat.com>
This patch adapts the mr creation/deletion code to be able to work with
any given mr struct pointer. All the APIs are adapted to take an extra
parameter for the mr.
mlx5_vdpa_create/delete_mr doesn't need a ASID parameter anymore. The
check is done in the caller instead (mlx5_set_map).
This change is needed for a followup patch which will introduce an
additional mr for the vq descriptor data.
Signed-off-by: Dragos Tatulea <dtatulea@nvidia.com>
Acked-by: Eugenio Pérez <eperezma@redhat.com>
Acked-by: Jason Wang <jasowang@redhat.com>
Message-Id: <20231018171456.1624030-12-dtatulea@nvidia.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
Reviewed-by: Si-Wei Liu <si-wei.liu@oracle.com>
Tested-by: Si-Wei Liu <si-wei.liu@oracle.com>
Tested-by: Lei Yang <leiyang@redhat.com>
The handling of the cvq iotlb is currently coupled with the creation
and destruction of the hardware mkeys (mr).
This patch moves cvq iotlb handling into its own function and shifts it
to a scope that is not related to mr handling. As cvq handling is just a
prune_iotlb + dup_iotlb cycle, put it all in the same "update" function.
Finally, the destruction path is handled by directly pruning the iotlb.
After this move is done the ASID mr code can be collapsed into a single
function.
Acked-by: Jason Wang <jasowang@redhat.com>
Acked-by: Eugenio Pérez <eperezma@redhat.com>
Signed-off-by: Dragos Tatulea <dtatulea@nvidia.com>
Message-Id: <20231018171456.1624030-8-dtatulea@nvidia.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
Reviewed-by: Si-Wei Liu <si-wei.liu@oracle.com>
Tested-by: Si-Wei Liu <si-wei.liu@oracle.com>
Tested-by: Lei Yang <leiyang@redhat.com>
A firmware error is triggered when configuring a 9k MTU on the PF after
switching to switchdev mode and then using a vdpa device with larger
(1k) rings:
mlx5_cmd_out_err: CREATE_GENERAL_OBJECT(0xa00) op_mod(0xd) failed, status bad resource(0x5), syndrome (0xf6db90), err(-22)
This is due to the fact that the hw VQ size parameters are computed
based on the umem_1/2/3_buffer_param_a/b capabilities and all
device capabilities are read only when the driver is moved to switchdev mode.
The problematic configuration flow looks like this:
1) Create VF
2) Unbind VF
3) Switch PF to switchdev mode.
4) Bind VF
5) Set PF MTU to 9k
6) create vDPA device
7) Start VM with vDPA device and 1K queue size
Note that setting the MTU before step 3) doesn't trigger this issue.
This patch reads the forementioned umem parameters at the latest point
possible before the VQs of the device are created.
v2:
- Allocate output with kmalloc to reduce stack frame size.
- Removed stable from cc.
Fixes: 1a86b377aa ("vdpa/mlx5: Add VDPA driver for supported mlx5 devices")
Signed-off-by: Dragos Tatulea <dtatulea@nvidia.com>
Message-Id: <20230831155702.1080754-1-dtatulea@nvidia.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
Acked-by: Jason Wang <jasowang@redhat.com>
The error path in setup_driver deletes the debugfs entry but doesn't
clear the pointer. During .dev_del the invalid pointer will be released
again causing a crash.
This patch fixes the issue by always clearing the debugfs entry in
mlx5_vdpa_remove_debugfs. Also, stop removing the debugfs entry in
.dev_del op: the debugfs entry is already handled within the
setup_driver/teardown_driver scope.
Cc: stable@vger.kernel.org
Fixes: f0417e72ad ("vdpa/mlx5: Add and remove debugfs in setup/teardown driver")
Signed-off-by: Dragos Tatulea <dtatulea@nvidia.com>
Reviewed-by: Gal Pressman <gal@nvidia.com>
Message-Id: <20230829174014.928189-2-dtatulea@nvidia.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
Acked-by: Jason Wang <jasowang@redhat.com>
Pull virtio updates from Michael Tsirkin:
"A small pull request this time around, mostly because the vduse
network got postponed to next relase so we can be sure we got the
security store right"
* tag 'for_linus' of git://git.kernel.org/pub/scm/linux/kernel/git/mst/vhost:
virtio_ring: fix avail_wrap_counter in virtqueue_add_packed
virtio_vdpa: build affinity masks conditionally
virtio_net: merge dma operations when filling mergeable buffers
virtio_ring: introduce dma sync api for virtqueue
virtio_ring: introduce dma map api for virtqueue
virtio_ring: introduce virtqueue_reset()
virtio_ring: separate the logic of reset/enable from virtqueue_resize
virtio_ring: correct the expression of the description of virtqueue_resize()
virtio_ring: skip unmap for premapped
virtio_ring: introduce virtqueue_dma_dev()
virtio_ring: support add premapped buf
virtio_ring: introduce virtqueue_set_dma_premapped()
virtio_ring: put mapping error check in vring_map_one_sg
virtio_ring: check use_dma_api before unmap desc for indirect
vdpa_sim: offer VHOST_BACKEND_F_ENABLE_AFTER_DRIVER_OK
vdpa: add get_backend_features vdpa operation
vdpa: accept VHOST_BACKEND_F_ENABLE_AFTER_DRIVER_OK backend feature
vdpa: add VHOST_BACKEND_F_ENABLE_AFTER_DRIVER_OK flag
vdpa/mlx5: Remove unused function declarations
We were allocating irq vectors at the time the aux dev was probed,
but that is before the PCI VF is assigned to a separate iommu domain
by vhost_vdpa. Because vhost_vdpa later changes the iommu domain the
interrupts do not work.
Instead, we can allocate the irq vectors later when we see DRIVER_OK and
know that the reassignment of the PCI VF to an iommu domain has already
happened.
Fixes: 151cc834f3 ("pds_vdpa: add support for vdpa and vdpamgmt interfaces")
Signed-off-by: Allen Hubbe <allen.hubbe@amd.com>
Signed-off-by: Shannon Nelson <shannon.nelson@amd.com>
Acked-by: Jason Wang <jasowang@redhat.com>
Message-Id: <20230711042437.69381-5-shannon.nelson@amd.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
Our driver sets a mac if the HW is 00:..:00 so we need to be sure to
advertise VIRTIO_NET_F_MAC even if the HW doesn't. We also need to be
sure that virtio_net sees the VIRTIO_NET_F_MAC and doesn't rewrite the
mac address that a user may have set with the vdpa utility.
After reading the hw_feature bits, add the VIRTIO_NET_F_MAC to the driver's
supported_features and use that for reporting what is available. If the
HW is not advertising it, be sure to strip the VIRTIO_NET_F_MAC before
finishing the feature negotiation. If the user specifies a device_features
bitpattern in the vdpa utility without the VIRTIO_NET_F_MAC set, then
don't set the mac.
Fixes: 151cc834f3 ("pds_vdpa: add support for vdpa and vdpamgmt interfaces")
Signed-off-by: Shannon Nelson <shannon.nelson@amd.com>
Message-Id: <20230711042437.69381-3-shannon.nelson@amd.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
Acked-by: Jason Wang <jasowang@redhat.com>
mlx5_vdpa_destroy_mr can be called from .set_map with data ASID after
the control virtqueue ASID iotlb has been populated. The control vq
iotlb must not be cleared, since it will not be populated again.
So call the ASID aware destroy function which makes sure that the
right vq resource is destroyed.
Fixes: 8fcd20c307 ("vdpa/mlx5: Support different address spaces for control and data")
Signed-off-by: Eugenio Pérez <eperezma@redhat.com>
Reviewed-by: Gal Pressman <gal@nvidia.com>
Message-Id: <20230802171231.11001-5-dtatulea@nvidia.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
Acked-by: Jason Wang <jasowang@redhat.com>
The mr->initialized flag is shared between the control vq and data vq
part of the mr init/uninit. But if the control vq and data vq get placed
in different ASIDs, it can happen that initializing the control vq will
prevent the data vq mr from being initialized.
This patch consolidates the control and data vq init parts into their
own init functions. The mr->initialized will now be used for the data vq
only. The control vq currently doesn't need a flag.
The uninitializing part is also taken care of: mlx5_vdpa_destroy_mr got
split into data and control vq functions which are now also ASID aware.
Fixes: 8fcd20c307 ("vdpa/mlx5: Support different address spaces for control and data")
Signed-off-by: Dragos Tatulea <dtatulea@nvidia.com>
Reviewed-by: Eugenio Pérez <eperezma@redhat.com>
Reviewed-by: Gal Pressman <gal@nvidia.com>
Message-Id: <20230802171231.11001-3-dtatulea@nvidia.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
Acked-by: Jason Wang <jasowang@redhat.com>
The vdpa_nl_policy structure is used to validate the nlattr when parsing
the incoming nlmsg. It will ensure the attribute being described produces
a valid nlattr pointer in info->attrs before entering into each handler
in vdpa_nl_ops.
That is to say, the missing part in vdpa_nl_policy may lead to illegal
nlattr after parsing, which could lead to OOB read just like CVE-2023-3773.
This patch adds the missing nla_policy for vdpa max vqp attr to avoid
such bugs.
Fixes: ad69dd0bf2 ("vdpa: Introduce query of device config layout")
Signed-off-by: Lin Ma <linma@zju.edu.cn>
Cc: stable@vger.kernel.org
Message-Id: <20230727175757.73988-7-dtatulea@nvidia.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
The vdpa_nl_policy structure is used to validate the nlattr when parsing
the incoming nlmsg. It will ensure the attribute being described produces
a valid nlattr pointer in info->attrs before entering into each handler
in vdpa_nl_ops.
That is to say, the missing part in vdpa_nl_policy may lead to illegal
nlattr after parsing, which could lead to OOB read just like CVE-2023-3773.
This patch adds the missing nla_policy for vdpa queue index attr to avoid
such bugs.
Fixes: 13b00b1356 ("vdpa: Add support for querying vendor statistics")
Signed-off-by: Lin Ma <linma@zju.edu.cn>
Cc: stable@vger.kernelorg
Message-Id: <20230727175757.73988-5-dtatulea@nvidia.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
The vdpa_nl_policy structure is used to validate the nlattr when parsing
the incoming nlmsg. It will ensure the attribute being described produces
a valid nlattr pointer in info->attrs before entering into each handler
in vdpa_nl_ops.
That is to say, the missing part in vdpa_nl_policy may lead to illegal
nlattr after parsing, which could lead to OOB read just like CVE-2023-3773.
This patch adds the missing nla_policy for vdpa features attr to avoid
such bugs.
Fixes: 90fea5a800 ("vdpa: device feature provisioning")
Signed-off-by: Lin Ma <linma@zju.edu.cn>
Cc: stable@vger.kernel.org
Message-Id: <20230727175757.73988-3-dtatulea@nvidia.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
This commit enables the dynamic allocation of EQs at runtime, allowing
for more flexibility in managing completion EQs and reducing the memory
overhead of driver load. Whenever a CQ is created for a given vector
index, the driver will lookup to see if there is an already mapped
completion EQ for that vector, if so, utilize it. Otherwise, allocate a
new EQ on demand and then utilize it for the CQ completion events.
Add a protection lock to the EQ table to protect from concurrent EQ
creation attempts.
While at it, replace mlx5_vector2irqn()/mlx5_vector2eqn() with
mlx5_comp_eqn_get() and mlx5_comp_irqn_get() which will allocate an
EQ on demand if no EQ is found for the given vector.
Signed-off-by: Maher Sanalla <msanalla@nvidia.com>
Reviewed-by: Shay Drory <shayd@nvidia.com>
Reviewed-by: Moshe Shemesh <moshe@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
Pull virtio updates from Michael Tsirkin:
- resume support in vdpa/solidrun
- structure size optimizations in virtio_pci
- new pds_vdpa driver
- immediate initialization mechanism for vdpa/ifcvf
- interrupt bypass for vdpa/mlx5
- multiple worker support for vhost
- viirtio net in Intel F2000X-PL support for vdpa/ifcvf
- fixes, cleanups all over the place
* tag 'for_linus' of git://git.kernel.org/pub/scm/linux/kernel/git/mst/vhost: (48 commits)
vhost: Make parameter name match of vhost_get_vq_desc()
vduse: fix NULL pointer dereference
vhost: Allow worker switching while work is queueing
vhost_scsi: add support for worker ioctls
vhost: allow userspace to create workers
vhost: replace single worker pointer with xarray
vhost: add helper to parse userspace vring state/file
vhost: remove vhost_work_queue
vhost_scsi: flush IO vqs then send TMF rsp
vhost_scsi: convert to vhost_vq_work_queue
vhost_scsi: make SCSI cmd completion per vq
vhost_sock: convert to vhost_vq_work_queue
vhost: convert poll work to be vq based
vhost: take worker or vq for flushing
vhost: take worker or vq instead of dev for queueing
vhost, vhost_net: add helper to check if vq has work
vhost: add vhost_worker pointer to vhost_virtqueue
vhost: dynamically allocate vhost_worker
vhost: create worker at end of vhost_dev_set_owner
virtio_bt: call scheduler when we free unused buffs
...
This commit implements a better layout of the
live migration bar, therefore the accessors for virtqueue
state have been refactored.
This commit also add a comment to the probing-ids list,
indicating this driver drives F2000X-PL virtio-net
Signed-off-by: Zhu Lingshan <lingshan.zhu@intel.com>
Message-Id: <20230612151420.1019504-4-lingshan.zhu@intel.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
Pull mm updates from Andrew Morton:
- Yosry Ahmed brought back some cgroup v1 stats in OOM logs
- Yosry has also eliminated cgroup's atomic rstat flushing
- Nhat Pham adds the new cachestat() syscall. It provides userspace
with the ability to query pagecache status - a similar concept to
mincore() but more powerful and with improved usability
- Mel Gorman provides more optimizations for compaction, reducing the
prevalence of page rescanning
- Lorenzo Stoakes has done some maintanance work on the
get_user_pages() interface
- Liam Howlett continues with cleanups and maintenance work to the
maple tree code. Peng Zhang also does some work on maple tree
- Johannes Weiner has done some cleanup work on the compaction code
- David Hildenbrand has contributed additional selftests for
get_user_pages()
- Thomas Gleixner has contributed some maintenance and optimization
work for the vmalloc code
- Baolin Wang has provided some compaction cleanups,
- SeongJae Park continues maintenance work on the DAMON code
- Huang Ying has done some maintenance on the swap code's usage of
device refcounting
- Christoph Hellwig has some cleanups for the filemap/directio code
- Ryan Roberts provides two patch series which yield some
rationalization of the kernel's access to pte entries - use the
provided APIs rather than open-coding accesses
- Lorenzo Stoakes has some fixes to the interaction between pagecache
and directio access to file mappings
- John Hubbard has a series of fixes to the MM selftesting code
- ZhangPeng continues the folio conversion campaign
- Hugh Dickins has been working on the pagetable handling code, mainly
with a view to reducing the load on the mmap_lock
- Catalin Marinas has reduced the arm64 kmalloc() minimum alignment
from 128 to 8
- Domenico Cerasuolo has improved the zswap reclaim mechanism by
reorganizing the LRU management
- Matthew Wilcox provides some fixups to make gfs2 work better with the
buffer_head code
- Vishal Moola also has done some folio conversion work
- Matthew Wilcox has removed the remnants of the pagevec code - their
functionality is migrated over to struct folio_batch
* tag 'mm-stable-2023-06-24-19-15' of git://git.kernel.org/pub/scm/linux/kernel/git/akpm/mm: (380 commits)
mm/hugetlb: remove hugetlb_set_page_subpool()
mm: nommu: correct the range of mmap_sem_read_lock in task_mem()
hugetlb: revert use of page_cache_next_miss()
Revert "page cache: fix page_cache_next/prev_miss off by one"
mm/vmscan: fix root proactive reclaim unthrottling unbalanced node
mm: memcg: rename and document global_reclaim()
mm: kill [add|del]_page_to_lru_list()
mm: compaction: convert to use a folio in isolate_migratepages_block()
mm: zswap: fix double invalidate with exclusive loads
mm: remove unnecessary pagevec includes
mm: remove references to pagevec
mm: rename invalidate_mapping_pagevec to mapping_try_invalidate
mm: remove struct pagevec
net: convert sunrpc from pagevec to folio_batch
i915: convert i915_gpu_error to use a folio_batch
pagevec: rename fbatch_count()
mm: remove check_move_unevictable_pages()
drm: convert drm_gem_put_pages() to use a folio_batch
i915: convert shmem_sg_free_table() to use a folio_batch
scatterlist: add sg_set_folio()
...
Add support for generation of interrupts from the device directly to the
VM to the VCPU thus avoiding the overhead on the host CPU.
When supported, the driver will attempt to allocate vectors for each
data virtqueue. If a vector for a virtqueue cannot be provided it will
use the QP mode where notifications go through the driver.
In addition, we add a shutdown callback to make sure allocated
interrupts are released in case of shutdown to allow clean shutdown.
Signed-off-by: Eli Cohen <elic@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
Message-Id: <20230607190007.290505-1-dtatulea@nvidia.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
