Make mdev driver dma_unmap callback tolerant to unmaps come before device open
Subsystems (7)
v2 cover Make mdev driver dma_unmap callback tolerant to unmaps come before device open
Jason's "Connect VFIO to IOMMUFD" introduces vfio iommufd compat mode. Under this mode, vfio_iommufd_bind() creates an access which has an unmap callback, which can be called immediately. This means mdev drivers may receive unmap requests before the mdev is opened. For now, there are only three drivers (gvt, vfio-ap and vfio-ccw) providing dma_unmap(). vfio-ccw is fine with such requests. While gvt-g and vfio-ap may have potential problem with such requests due to internal implementation. This series tries to enhance the two drivers.
This series is based on Jason's below branch.
https://github.com/jgunthorpe/linux/tree/iommufd
(commit: 41973418f6c8c241ed5647d1408d5b917f24dfd8)
Change:
v2:
- Refine the cover letter and commit message of patch 0001 (Kevin)
- Rename patch 0001 to better fit the commit message
- Add r-b from Zhi for patch 0001
- tweak iova range test to assume page-aligned for patch 0002 (Jason)
- Remove break so all queues within range are removed for patch 0002 (Kevin)
v1: https://lore.kernel.org/kvm/[email protected]/
Regards,
Yi Liu
Matthew Rosato (1):
vfio/ap: validate iova during dma_unmap and trigger irq disable
Yi Liu (1):
i915/gvt: Move gvt mapping cache initialization to vGPU creation
drivers/gpu/drm/i915/gvt/gvt.h | 2 ++ drivers/gpu/drm/i915/gvt/kvmgt.c | 7 ++----- drivers/gpu/drm/i915/gvt/vgpu.c | 2 ++ drivers/s390/crypto/vfio_ap_ops.c | 18 +++++++++++++++++- 4 files changed, 23 insertions(+), 6 deletions(-)
Signature · 1 line
2.34.1
PATCH v2 1/2 i915/gvt: Move gvt mapping cache initialization to vGPU creation
vfio container registers .dma_unmap() callback after the device is opened. So it's fine for mdev drivers to initialize internal mapping cache in .open_device(). See vfio_device_container_register().
Now with iommufd an access ops with an unmap callback is registered when the device is bound to iommufd which is before .open_device() is called. This implies gvt's .dma_unmap() could be called before its internal mapping cache is initialized.
The fix is moving gvt mapping cache initialization to vGPU creation. While at it also move ptable initialization together.
--- drivers/gpu/drm/i915/gvt/gvt.h | 2 ++ drivers/gpu/drm/i915/gvt/kvmgt.c | 7 ++----- drivers/gpu/drm/i915/gvt/vgpu.c | 2 ++ 3 files changed, 6 insertions(+), 5 deletions(-)
drivers/gpu/drm/i915/gvt/gvt.h +2 -0
@@ -754,6 +754,8 @@ void intel_gvt_debugfs_remove_vgpu(struct intel_vgpu *vgpu);void intel_gvt_debugfs_init (struct intel_gvt *gvt);void intel_gvt_debugfs_clean (struct intel_gvt *gvt); +void gvt_cache_init (struct intel_vgpu *vgpu); +void kvmgt_protect_table_init (struct intel_vgpu *info);int intel_gvt_page_track_add (struct intel_vgpu *info,u64 gfn);int intel_gvt_page_track_remove (struct intel_vgpu *info,u64 gfn);int intel_gvt_dma_pin_guest_page (struct intel_vgpu *vgpu,dma_addr_t dma_addr);
drivers/gpu/drm/i915/gvt/kvmgt.c +2 -5
@@ -322,7 +322,7 @@ static void gvt_cache_destroy(struct intel_vgpu *vgpu) } } -static void gvt_cache_init (struct intel_vgpu *vgpu) +void gvt_cache_init (struct intel_vgpu *vgpu) { vgpu->gfn_cache =RB_ROOT ; vgpu->dma_addr_cache =RB_ROOT ;
@@ -330,7 +330,7 @@ static void gvt_cache_init(struct intel_vgpu *vgpu)mutex_init (&vgpu->cache_lock); } -static void kvmgt_protect_table_init (struct intel_vgpu *info) +void kvmgt_protect_table_init (struct intel_vgpu *info) {hash_init (info->ptable); }
@@ -671,9 +671,6 @@ static int intel_vgpu_open_device(struct vfio_device *vfio_dev) vgpu->attached = true; -kvmgt_protect_table_init (vgpu); -gvt_cache_init (vgpu); - vgpu->track_node.track_write = kvmgt_page_track_write; vgpu->track_node.track_flush_slot = kvmgt_page_track_flush_slot; kvm_page_track_register_notifier(vgpu->vfio_device.kvm,
drivers/gpu/drm/i915/gvt/vgpu.c +2 -0
@@ -382,6 +382,8 @@ int intel_gvt_create_vgpu(struct intel_vgpu *vgpu,intel_gvt_update_reg_whitelist (vgpu);mutex_unlock (&gvt->lock); +kvmgt_protect_table_init (vgpu); +gvt_cache_init (vgpu);return 0 ; out_clean_sched_policy:
Signature · 1 line
2.34.1
PATCH v2 2/2 vfio/ap: validate iova during dma_unmap and trigger irq disable
In reply to Yi LiuFrom: Matthew Rosato <[email protected]>
Currently, each mapped iova is stashed in its associated vfio_ap_queue; when we get an unmap request, validate that it matches with one or more of these stashed values before attempting unpins.
Each stashed iova represents IRQ that was enabled for a queue. Therefore, if a match is found, trigger IRQ disable for this queue to ensure that underlying firmware will no longer try to use the associated pfn after the page is unpinned. IRQ disable will also handle the associated unpin.
--- drivers/s390/crypto/vfio_ap_ops.c | 18 +++++++++++++++++- 1 file changed, 17 insertions(+), 1 deletion(-)
drivers/s390/crypto/vfio_ap_ops.c +17 -1
@@ -1535,13 +1535,29 @@ static int vfio_ap_mdev_set_kvm(struct ap_matrix_mdev *matrix_mdev,return 0 ; } +static void unmap_iova (struct ap_matrix_mdev *matrix_mdev,u64 iova,u64 length) +{ +struct ap_queue_table *qtable = &matrix_mdev->qtable; +struct vfio_ap_queue *q; +int loop_cursor; + +hash_for_each (qtable->queues, loop_cursor, q, mdev_qnode) { +if (q->saved_iova >= iova && q->saved_iova < iova + length) +vfio_ap_irq_disable (q); + } +} +static void vfio_ap_mdev_dma_unmap (struct vfio_device *vdev,u64 iova,u64 length) {struct ap_matrix_mdev *matrix_mdev = container_of(vdev, struct ap_matrix_mdev, vdev); -vfio_unpin_pages (&matrix_mdev->vdev, iova,1 ); +mutex_lock (&matrix_dev->mdevs_lock); + +unmap_iova (matrix_mdev, iova, length); + +mutex_unlock (&matrix_dev->mdevs_lock); }/**
Signature · 1 line
2.34.1
Quoted 9 lines · Yi Liu
Jason's "Connect VFIO to IOMMUFD" introduces vfio iommufd compat mode. Under this mode,
vfio_iommufd_bind()creates an access which has an unmap callback, which can be called immediately. This means mdev drivers may receive unmap requests before the mdev is opened. For now, there are only three drivers (gvt, vfio-ap and vfio-ccw) providingdma_unmap(). vfio-ccw is fine with such requests. While gvt-g and vfio-ap may have potential problem with such requests due to internal implementation. This series tries to enhance the two drivers.
Quoted 5 lines
This series is based on Jason's below branch.
https://github.com/jgunthorpe/linux/tree/iommufd
(commit:
41973418f6c8c241ed5647d1408d5b917f24dfd8)
it's resent as below link. Please ignore this series.
https://lore.kernel.org/kvm/[email protected]/
Quoted 34 lines
Change:
v2:
- Refine the cover letter and commit message of patch 0001 (Kevin)
- Rename patch 0001 to better fit the commit message
- Add r-b from Zhi for patch 0001
- tweak iova range test to assume page-aligned for patch 0002 (Jason)
- Remove break so all queues within range are removed for patch 0002 (Kevin)v1: https://lore.kernel.org/kvm/[email protected]/
Cc Tony Krowiak <[email protected]>Cc Halil Pasic <[email protected]>Cc Jason Herne <[email protected]>Cc Zhenyu Wang <[email protected]>Cc Zhi Wang <[email protected]>Cc Kevin Tian <[email protected]>Regards,
Yi LiuMatthew Rosato (1):
vfio/ap: validate iova duringdma_unmapand trigger irq disableYi Liu (1):
i915/gvt: Move gvt mapping cache initialization to vGPU creationdrivers/gpu/drm/i915/gvt/gvt.h | 2 ++ drivers/gpu/drm/i915/gvt/kvmgt.c | 7 ++----- drivers/gpu/drm/i915/gvt/vgpu.c | 2 ++ drivers/s390/crypto/vfio_ap_ops.c | 18 +++++++++++++++++- 4 files changed, 23 insertions(+), 6 deletions(-)
Signature · 2 lines
Regards, Yi Liu