When a PASID is stopped or terminated, there can be pending PRQs (requests that haven't received responses) in the software and remapping hardware. The pending page requests must be drained so that the pasid could be reused. The chapter 7.10 in the VT-d specification specifies the software steps to drain pending page requests and responses.
This includes two parts: - PATCH 1/4 ~ 2/4: refactor the qi_submit_sync() to support multiple descriptors per submission which will be used in the following patch. - PATCH 3/4 ~ 4/4: add page request drain support after a pasid entry is torn down. Best regards, baolu Change log: v3->v4: - Remove prq drain in mm notifier; - Set PASID FPD bit when pasid is cleared in mm notifier and clear it in unbound(). v2->v3: - Address Kevin's review comments - Squash the first 2 patches together; - The prq thread is serialized, no need to consider reentrance; - Ensure no new-coming prq before drain prq in queue; - Handle page request overflow case. v1->v2: - Fix race between multiple prq handling threads. Lu Baolu (5): iommu/vt-d: Multiple descriptors per qi_submit_sync() iommu/vt-d: debugfs: Add support to show inv queue internals iommu/vt-d: Disable non-recoverable fault processing before unbind iommu/vt-d: Add page request draining support iommu/vt-d: Remove redundant IOTLB flush drivers/iommu/dmar.c | 63 ++++++++------ drivers/iommu/intel-iommu-debugfs.c | 62 ++++++++++++++ drivers/iommu/intel-iommu.c | 4 +- drivers/iommu/intel-pasid.c | 30 +++++-- drivers/iommu/intel-pasid.h | 3 +- drivers/iommu/intel-svm.c | 123 ++++++++++++++++++++++++---- drivers/iommu/intel_irq_remapping.c | 2 +- include/linux/intel-iommu.h | 13 ++- 8 files changed, 247 insertions(+), 53 deletions(-) -- 2.17.1 _______________________________________________ iommu mailing list iommu@lists.linux-foundation.org https://lists.linuxfoundation.org/mailman/listinfo/iommu